DEVELOPING

LANGUAGE
TEACHER
AUTONOMY
THROUGH

ACTION
RESEARCH
KENAN DIKILITAŞ AND
CAROL GRIFFITHS
 Developing Language Teacher Autonomy
through Action Research

“Helping teachers engage autonomously in action research is a very worthwhile enterprise.

Beneficiaries are likely to include learners, schools and their cultures, as well as the teachers
themselves. Getting started is not always easy, though, and, with valuable examples, this vol-
ume takes teachers new to action research through crucial steps.”
—Mark Wyatt,
Senior Lecturer, University of Portsmouth, UK

“For many practitioners Action Research remains at the conceptual level; they want to be
involved, yet are not sure of how to implement it in their own setting. This book is of great
value to practitioners eager to be involved in action research. Each step is illustrated with
examples from real studies, from the very first steps finding a question to the start of the next
research cycle. This teacher-friendly guide is sure to be an important source for any teacher,
new or experienced, who is interested in professional development.”
—Derin Atay,
Professor, Bahçeşehir University, Turkey
 Kenan Dikilitaş • Carol Griffiths

Developing
Language Teacher
Autonomy through
Action Research
Kenan Dikilitaş Carol Griffiths
Faculty of Educational Sciences, Freelance,
Bahçeşehir University, Istanbul, Turkey
Istanbul, Turkey

ISBN 978-3-319-50738-5    ISBN 978-3-319-50739-2 (eBook)

DOI 10.1007/978-3-319-50739-2

Library of Congress Control Number: 2016963727

© The Editor(s) (if applicable) and The Author(s) 2017

This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether
the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of
illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and trans-
mission or information storage and retrieval, electronic adaptation, computer software, or by similar or
dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or
the editors give a warranty, express or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.

Cover illustration: ©Design Pics Inc/Getty

Printed on acid-free paper

This Palgrave Macmillan imprint is published by Springer Nature

The registered company is Springer International Publishing AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
I dedicate this book to my late father, and to my son, Derin, and my wife,
Derya. They have always been supportive of my work. (Kenan)
And I would like to dedicate this book to my parents, Ben and Ranee, now
both deceased, who always had a vision beyond the background to which
they were born. (Carol)
 Foreword

This book is an outcome of intensive engagement in facilitating and sup-

porting teachers who are engaged in action research at different higher
education institutes in Turkey. The idea of teachers being engaged in
action research, which gained momentum in 2010 with initiatives in
Izmir, is relatively new in Turkey. I will describe how I began this action
research mentor journey mainly because it could be useful for readers to
know how the idea of writing a book on action research emerged. The
story of action research engagement in Turkey is also documented in sev-
eral previous publications (see Çelik and Dikilitaş 2015; Dikilitaş 2015;
Dikilitaş and Mumford 2016; Smith 2015; Wyatt and Dikilitaş 2015).
When I took up the position at a university’s preparatory English
language school as a teacher trainer in 2010, I was supposed to work
with almost 40 teachers to promote and support their professional devel-
opment. I thought about several options for professional development
activities, but I wanted to make one central because I believed that the
project should set goals for teachers, ensure longer engagement, inspire
personal active engagement, and offer an outcome such as dissemination
through presenting, writing, or publishing. Classroom observation was
one good option but it would not be central as it might not address some
of my own expectations such as writing and publishing. I also thought
about running workshops on specific pedagogical issues, but then I
would have to take too much control myself and minimize the ­teachers’
vii
viii  Foreword

active engagement in learning. As an emerging researcher who was just

starting to write my PhD dissertation, I thought of action research as
a professional development strategy which could address most of my
expectations from teachers’ engagement in professional development. As
I previously reported in an interview with Richard Smith in 2012, who
later published it in 2014, my idea was to initiate action research as a
professional development tool because it was compatible with my own
overall learning experience and because I believed teachers need not be
fed with knowledge but rather discover relevant knowledge themselves.
In the first year (2010), there were several challenges in supporting
teachers, such as teachers’ own lack of knowledge about action research
and of research in general, as they reported several times. Engagement in
action research was not in line with what they would expect from profes-
sional development activities as they were used to lecture-style activities
in which they took part as a recipient of knowledge rather than a creator
of it. As a new way of learning, action research seemed to many of these
teachers to be challenging due not only to lack of research knowledge but
also to contextual constraints such as limited time to devote to doing and
sustaining research. Therefore, it took time to explain to them what kind
of research they would undertake and how they were going to conduct it.
In the first year, action research engagement was a top-down decision by
the administration and me as opposed to traditional transmission-based
learning as teachers expected.
To address these concerns I started to run sessions telling them
more about what action research is and what actual steps and cycles
are involved in doing it. We had discussions about classroom issues
in groups, individually, and even sometimes in pairs. By starting to
talk about classroom issues and planning how they could investigate
them, the teachers developed a sense of self-efficacy in starting their
own research, though not all of them were attracted to it and confident
enough with the idea. In the initial years, I provided more structured
support with explicit guidance on planning action research. I gave them
a proposal guideline which included questions such as title, purpose,
participants, data collection tools, and their expectations from doing
research as documented in Çelik and Dikilitaş (2015). Such tools facili-
tated some teachers’ initiation and helped them believe they could do
  Foreword 
   ix

it. I then continued to provide systematic support for those who were
determined to proceed. We had weekly focus group meetings where
each action researcher presented a proposal to get feedback from col-
leagues, which also gave them confidence to discuss research issues. In
one-on-one meetings we discussed relevant articles to develop knowl-
edge about research issues, and planned classroom activities to collect
data or introduce new ways of teaching and to assess and reflect upon
the impact thereafter. I kept interaction and dialogue actively going in
order to help them complete their action research though some were
not able to do so due to limited commitment and high workload or
to personal dissatisfaction with the idea of learning and development
through action research. At the end of the first year, I held an in-house
event where those who successfully completed their research presented
it, followed by an edited book (Dikilitaş 2011).
In the second year (2011), the project developed. One development
was that teachers started to collaborate with one another and inves-
tigate shared concerns, and/or puzzles. Unlike the previous year, as a
trainer I found myself less dominant. Some developed confidence in
initiating and sustaining their research process with less and less depen-
dence on my support.There were also new teachers who started to
engage in action research in the second year. I provided more focused
and structured support for them, but they were also helped out by their
colleagues who had experienced action research the previous year. As
the teachers engaged in action research, they built rapport and worked
more in collaboration, which also led them to becoming more autono-
mous and self-confident.
In 2012, the project continued with more enthusiasm and I had the
opportunity to work with external mentors such as Simon Borg and
Richard Smith, who were invited to my institution to work with teachers
who were doing action research during the projects. Their feedback on
the teachers’ action research and comments for the future of the project
were valuable as deeper insights emerged such as developing the idea of
presentation and publication, which would lead to changes in the way
presentations were delivered. They also attended the conference at the end
of the year as plenary speakers and attracted more audience from other
nearby institutions and cities (for further details, see Dikilitaş 2013).
x  Foreword

The project was supported in 2013 by the then coordinator of IATEFL

Research Special Interest group (ReSIG) Richard Smith, who met with
the 18 teachers and provided feedback during the year. Such external
help created motivation and enthusiasm for the teachers to sustain their
engagement in action research. Supported internally by me and exter-
nally by other well-known names, the teachers felt valued and became
more concentrated on their work. The feedback they received from me
and the external mentors included the methodological designs of action
research as well as the overall purpose and topic of research. Receiving
feedback from different mentors intensified the degree of learning and
development and provided motivation for the teachers to move on.
The teachers started to develop their understandings of action research
through dialogues with others rather than working within template pro-
posals and structured guidelines. The teachers stared to plan and develop
their own research engagement path and created for themselves unique
ways of researching.
At the end of the 2014 project, another annual conference was held
with the participation of Richard Smith, Anne Burns, Mark Wyatt,
and Judith Hanks, who gave enthusiastic talks and commented on the
development of action research in Turkey both in their speeches and in
the concurrent sessions. Their individual feedback in the sessions were
also valuable for the presenters. That year, the conference was called the
IATEFL ReSIG Teachers Research Conference, which attracted relatively
more presenters and attendees from Turkey and beyond. Another devel-
opment from the previous year was that the annual conference book was
edited by Kenan Dikilitaş, Richard Smith, and Wayne Trotman (2015)
and published as an IATEFL ReSIG book, which was made open access
in the official website.
The same conference was held in 2015 and 2016 in collaboration with
the IATEFL ReSIG in İzmir and in İstanbul, respectively. The grow-
ing number of attendees—92  in 2015 and 120  in 2016—indicated a
developing interest in action research particularly in Turkey. The follow-
­up publication of the 2015 conference in İzmir was edited by Kenan
Dikilitaş, Mark Wyatt, Judith Hanks, and Deborah Bullock, which is
now open access in the website of IATEFL ReSIG. An initiation which
began in 2010 with a very small scale project developed into an inter-
  Foreword 
   xi

national one with the external supporters at different stages ranging

from helping action researchers in process to contributing to the annual
­conferences and publications.
It is clear that the project has developed in different ways. For exam-
ple, it has become an international event which brings together teachers,
educators, and trainers who are interested in action research. This is one
of the major sources of motivation for teachers to engage in research
and complete their work by the end of the academic year in order to be
able to present it. Another development has been the liberation of the
teachers in taking the control of their own research often moving to rela-
tively less interference by the mentors and more teacher autonomy.The
teachers have started to select topics to research and conduct research in
their own classrooms with more collaboration and individual efforts. The
degree of autonomy which has developed throughout successive projects
with the active engagement of teachers has manifested itself at different
levels, such as developing skills for reading articles critically, collecting
and analysing data, reflecting on the findings, presenting and writing
research. The observed teacher autonomy development was triggered by
the sustained engagement in doing, presenting, and writing up research,
which was well supported by the ongoing collaborative work, the fol-
low-­up conferences, and publications. This clearly showed the positive
impact of the research engagement experiences, teacher autonomy, and
demonstrated sustainability. The teachers learnt to research not only by
monitoring their own development and learning in the successive years
but also by assessing others’ work and receiving constructive rather than
judgmental feedback during the conference and the editorial process of
publication.
I have also developed a sense of autonomy in mentoring action
researchers as a result of the successive experiences in these confer-
ences and publications, which had an immense effect in cultivating my
own professional development as a research mentor. Over the years,
I have taken up different mentor roles because I also developed my
skills as a mentor, which led me to taking on varying roles. Malderez
and Bodóczky (1999) list the roles of mentors as model, acculturator,
support, sponsor, and educator, whereas Halai (2006) categorizes these
xii  Foreword

roles into four: an expert coach, a subject specialist, a critical friend,

and a learner. I ­recognize that I have fulfilled almost all of these roles
while mentoring teachers who were doing action research.
–– Doing research with the teachers or even sometimes presenting with
them as a model (Malderez and Bodóczky 1999)
–– providing workshops as a subject specialist (Halai 2006), and as an
educator (Malderez and Bodóczky 1999)
–– providing opportunities for collaborative reflection (Malderez and
Bodóczky 1999) in the expert coach role (Halai 2006)
–– developing relationships—critical friend (Child and Merrill 2003;
Halai 2006)
–– collaborating on specific research issues, as a mentor as learner (Halai
2006)
–– taking on the role of support (Malderez and Bodóczky 1999)
through management and leadership
–– acculturator by introducing them into the research community
(Hobson and Sharp 2005) through publication opportunities in the
follow-up books.
–– sponsor holding conferences and facilitating opportunities for pub-
lications and bringing in people for them to discuss with (Malderez
and Bodóczky 1999)
Apart from these roles, I tried to create learning opportunities (Orland-­
Barak and Rachamim 2009) and to sustain interaction and support prac-
tical knowledge growth (Wyatt and Arnold 2012) by observing and giving
feedback on action research.
Participant-sensitive mentoring involves considering each researcher
or research group with their unique needs and addressing them in a con-
structive way. The degree of sustainability of these successive projects and
publications spread across Turkey has been intensified by the affordances
of the resourceful and supportive environment created by internal and
external mentors.
Teacher autonomy development from doing action research, which is
the focus of this book, is the main purpose of research mentoring. While
I developed several research mentoring skills as outlined above, I also had
the opportunity to integrate them into the support and facilitation I was
  Foreword 
   xiii

offering to the action researchers. Challenged by the idea that action

research was new in the context in which I was working, I brought out
different strategies to overcome them, which spurred me to renew my
approaches to mentoring each action researcher. It was the close work-
ing opportunity with the teachers that facilitated the interaction and the
process of support and collaboration. As a mentor, I have accumulated
over the years information about the researchers, their classroom teach-
ing practices, and the way they engage in research. In many ways such
opportunities served as facilitative factors for my mentoring and for the
action researchers’ learning opportunities to investigate their practices
and develop gradually increasing teacher autonomy to do research both
independently and collaboratively in the context. The sustained research
mentoring and ongoing engagement in doing research helped educate
action researchers who were committed to increasing their professional
engagement in their careers.
Joint authorship is a challenging process especially when the authors
live in different cities. However, the authors of this volume tried to keep
to the deadlines and exchanged a number of emails and had face-to-face
meetings (though limited) in order to maximize the potential to con-
tribute to the book. The process of writing through collaboration was
managed by the authors writing different chapters and sharing with each
other for further revision and editing. The chapters were peer-reviewed
and improved through negotiation and agreement. We also discussed
sequencing chapters, sorting out chapter organization, and selecting titles
and sub-titles.
In this book we have documented the ways in which teachers can
engage in action research and develop as professional teachers committed
to an inquiry stance in generating personal pedagogical knowledge and
developing autonomy. Equally importantly, this book is a source which
teacher educators who wish to support teachers doing action research can
employ in their mentoring process.

Kenan Dikilitaş
 Preface: Overview

How Is This Book Different?

1. It aims to develop teacher autonomy by means of engagement in
action research.
2. Although acknowledging that the research process is not always linear,
the book aims to set out the action research process in a clear progres-
sion which teachers can adapt to their needs.
3. The book aims to get teachers to engage with the process by means of
examples, narratives, questions, and tasks.
4. Teachers who follow these tasks might be able to have a complete
research project by the time they have finished working through the
book. This means that the work they do will have a tangible and usable
outcome.
5. Examples of actual action research project material will be included in
order to provide teachers with concrete models which they might
examine, adapt to their own needs, and, where appropriate, replicate.
6. Teacher narratives from those who have engaged in the action research
process are included to add a “human” touch, hopefully to inspire
those who need encouragement, and to reassure those who are experi-
encing problems that they are not the only ones having these kinds of
difficulties.

xv
xvi  Preface: Overview

7. This book is designed to be highly practical. It gives multiple ideas for

how to establish research questions, to choose appropriate methodolo-
gies, to adapt to the existing context, and to go about collecting data.
It provides examples of possible instruments, clear instructions for
how to carry out the most common kinds of statistical procedures,
ideas for how to do presentations and write up research findings for
those who want to do this. All of these steps can be very intimidating
for a new researcher (or even, sometimes, more experienced ones!).
8. In spite of the practical bias, the book aims to be theoretically and
ethically rigorous, and presents theoretically and ethically sound
methodologies throughout.
9. The book also has an extensive glossary which explains in some detail
the terms used in the book. Since this topic requires some quite spe-
cialized vocabulary (“jargon”, if you like), which can be intimidating
for non-experts (most of us!), a glossary which can provide quick and
easy reference is a major asset.

Audience
1. Trainee teachers who are working for a Diploma and who have access
to a class (maybe practice teaching) where they can conduct the
research.
2. In-service teachers who want to expand their own professional hori-
zons and develop their autonomy.
3. In-service teachers who are working for a higher qualification (e.g. a
Master’s or PhD).
4. Teacher educators who are working with in-service or trainee teachers
who wish to guide their students through the process of action
research.
 Questions to Consider

The questions are designed to revise the contents of the chapters and to
create discussion and reflection. They may be useful for individuals who
want to check their understanding of what they have read, for groups
which may benefit from some prompt to stimulate discussion, or for
teacher trainers who might use them to stimulate class discussion or as
the basis for assignment work.

Tasks
The tasks are designed to be cumulative, so that, if teachers wish, they
could be combined to form a complete research project by the end of the
book. This means that, instead of having just more piles of assignments
(as we all have had), the progress through this book has a tangible out-
come, which may be satisfying in itself, or which may be used as the basis
for a presentation or for publication.

xvii
 Acknowledgements

Thanks to those who have contributed studies or narratives to the book,

which adds variety, exemplification, and a “human” element.
Thanks to knowledgeable individuals (e.g. Richard Smith, Mark Wyatt
and Anne Burns, İsmail Hakkı Erten, Judith Hanks, Derin Atay, Şirin
Karadeniz, Sinemis Vatanartıran, Olcay Sert, Demet Yaylı, Enisa Mede,
Hatime Çiftçi, Aylin Tekiner Tolu, Yeşim Keşli Dollar, Servet Çelik,
Koray Haki Akyazı, Kerim Biçer) for their advice.
Thanks to colleagues for their encouragement and support, to family
and friends for their patience, and to students for their goodwill.

xix
 Contents

1 Introduction to Action Research    1

2 Clarifying the Research Questions or Hypotheses   45

3 Research Methods: Options and Issues   61

4 Thinking About the Context: Setting (Where?) and 

Participants (Who?)  89

5 Collecting the Data 107

6 Analysing the Data  129

7 Discussing the Data  161

8 Example Studies  171

9 Presenting and Writing Up Action Research    217

xxi
xxii  Contents

10 Researcher Narratives   237

11 Conclusion   263

Appendix: A Mixed Methods Study   273

Appendix: Classroom Culture Questionnaire   281

Glossary   283

References   293

Index   301
 List of Figures

Fig. 1.1 Possible conditions for collaboration 14

Fig. 1.2 Stages of AR 20
Fig. 6.1 Sample SPSS spreadsheet 133
Fig. 6.2 Sample reliability statistics 134
Fig. 6.3 Grid for reliability if item deleted 135
Fig. 6.4 Sample factor analysis grid (three-factor solution) 136
Fig. 6.5 Sample one-factor solution grid 137
Fig. 6.6 Sample tests for normality of distribution 138
Fig. 6.7 Sample frequency matrix 140
Fig. 6.8 Sample median matrix 141
Fig. 6.9 Sample mode matrix 142
Fig. 6.10 Sample Spearman’s correlation matrix 144
Fig. 6.11 Sample matrix for Mann–Whitney U test of
difference—gender146
Fig. 6.12 Sample matrix for Mann–Whitney U test of
difference—nationality147
Fig. 6.13 Sample mean ranks grid 148
Fig. 6.14 Teacher’s error correction strategies 152
Fig. 9.1 Sample poster (http://rukie.edu.glogster.com
fostering-speaking/with Rukiye’s permission) 220

xxiii
 List of Tables

Table 1.1 Examples of PEPAs 29

Table 1.2 Forms of AR 34
Table 3.1 Comparison of quantitative and qualitative research
paradigms62
Table 3.2 Points to consider when constructing questionnaires 68
Table 3.3 Reading strategy questionnaire 69
Table 3.4 Which of the following do you normally do when you
learn vocabulary? Please tick the top three 70
Table 3.5 Grammar learning preferences 71
Table 3.6 Please tick the option you think is most appropriate for
yourself and give a reason 72
Table 3.7 Sample interview schedule 78
Table 4.1 Contextual factors to be considered when planning a
study100
Table 5.1 Data sources to address research questions 113
Table 5.2 Interview stages with sample questions 124
Table 5.3 Question types with examples 124
Table 5.4 Sample observation questions 125
Table 6.1 Approximate effect size thresholds for Eta values 150
Table 6.2 Characteristics of deductive and inductive approaches
to data analysis 151
Table 6.3 Median ratings for impolite classroom behaviour 155

xxv
xxvi  List of Tables

Table 8.1 Group statistics of pre-test total scores 179

Table 8.2 Group statistics of post-test total scores 180
Table 8.3 Strategy frequency medians for males, females, and
overall208
Table 8.4 Significant correlations between pass–fail grades and
strategy items 209
Table 9.1 Data collection tools 228
Table 9.2 Major sections of a research article 231
Table 9.3 Informal research organization 232
Table 9.4 Benefits of publishing action research, summarized from
Burns (2014) 233
Table A.1 Median ratings of questionnaire items (the higher the
rating, the more impolite the behaviour is considered to be) 276
 1
Introduction to Action Research

What Is Action Research?

1. Since there is “action”, this implies that there will be some kind of active
investigation of whatever is seen to be the problem to be fixed, the puzzle
to be considered, the question to be answered, or the issue to be addressed.
2. Since there is “action”, there must be an agent, that is, someone per-
forming the action. In the case of action research (AR), this needs to
be someone who is in some way a participant in the context where the
problem/question/issue/puzzle is to be investigated. For the purposes
of this book, although much of it may be applied more widely (e.g. to
medical or workplace environments), we will focus on educational
contexts, and especially on the teacher and/or the students.

Background
Since the term was first used by Lewin (1946), AR has been defined
in various ways over the years. Nunan (1992) provides a “minimalist
definition, containing a question/issue, data, and interpretive analysis”

© The Author(s) 2017 1

K. Dikilitaş, C. Griffiths, Developing Language Teacher Autonomy
through Action Research, DOI 10.1007/978-3-319-50739-2_1
2  Developing Language Teacher Autonomy through Action Research

(p.  18), while Burns (2010) further explains that in AR “the teacher
becomes an ‘investigator’ or ‘explorer’ of his or her personal teaching
context, while at the same time being one of the participants in it”
(p. 2). AR has become one of the critical strategies that teachers have
engaged in for their professional development with a view to gain-
ing deeper insights into classroom contexts including learners, teach-
ing practices, and classroom management. Therefore, issues identified
by teachers as practitioners are of great importance in that they are
examples of the critical actual lived experiences that could inform poli-
cymakers, teacher-educators, or academic researchers. Teachers usually
identify issues that may contribute to their understanding and develop-
ment of pedagogic concerns including specific areas of teaching and
personal teaching strategies.
AR is an important strategy for professional development since it
allows teachers to create opportunities for developing awareness and
autonomy in teacher development. This is likely to be achieved by teach-
ers because AR encourages teachers to make decisions on salient issues,
to freely select the topic of research for their personal benefit, to identify
ways of developing a research plan, and to interpret the findings for their
own purposes. Such an insightful engagement in their own pedagogic
issues is critical in that

(a) it is a bottom-up rather than a top-down approach to address critical

aspects of teaching and learning.
(b) it has a natural relevance to the context in which teachers work.
(c) it allows for follow-up monitoring and impact of the developing
teaching and learning practices.
(d) it liberates teachers with a sense of agency and ownership to deal with
their own problems, critical questions, points to improve or puzzles,
thereby promoting teacher autonomy.

According to Burns (2010) AR often starts from something that teachers

have been puzzled, uncertain, or dissatisfied with for a while, a “burning
question” (p. 3) or issue they have always wanted to experiment with, a
change they would like to see happening in themselves or their learners,
or a desire to make a difference
 1  Introduction to Action Research 
   3

AR is so much embedded in classroom practices as part of professional

development that other benefits or influences on teachers often go unno-
ticed. For example, teachers who engage in AR might have instrumental
purposes such as using AR as a springboard to an academic career (e.g.
initiating a graduate degree) or to becoming a published teacher for get-
ting promotion. AR also recognizes and is built on the wealth of expe-
riential and practical knowledge which teachers have developed over the
years by means of immediate access to classroom opportunities.

Problem Solving or Problematizing

AR engagement might help teachers learn not only to problem-solve

but also to problematize (Burns and Richards 2012). One of the criti-
cal issues regarding AR is that it requires selecting a classroom problem
followed by reflecting, planning, acting, observing, re-reflecting, and re-­
planning classroom practices (Burns 2010). However, Burns (2015) also
uses the term “problematizing” to mean that AR is not only a tool to solve
pedagogic problems identified at the onset of research, but a means to
help develop alternative ways of thinking and improve practices. Burns
(2010) also implies that what is meant by “problem” may not necessarily
refer to a kind of deficiency in classroom teaching practices but a puzzling
issue. Though AR is seen as a problem-solving tool with engagement
in classroom-based research that integrates a new intervention into the
existing classroom practices through the use of teaching-oriented data
collection tools, it could help teachers understand the nature or root of
what they may call “problem”, thereby contributing to new insights and
approaches.

AR as a Cyclical and Spiral-Based Process

Burns (1999) sees AR as a cycle that is composed of four phases including

planning, acting, observing, and reflecting. Each cycle has a specific func-
tion in the researching process like a set of guidelines for teachers who
investigate their practices. We will discuss each cycle below to give an idea
of how they can be operationalized in AR.
4  Developing Language Teacher Autonomy through Action Research

Phase 1: Develop a Plan of Action to

(a) improve what is already happening or

(b) identify and examine a “puzzle” or problem area.

In the initial phase, Burns (2010) focuses on the identification of the

issue to be investigated. This phase includes working on identifying a
research focus, narrowing it down and developing research questions
as well as planning reliable and valid data collection tools, dealing with
ethical issues, and reviewing the relevant existing literature. This phase is
important for careful preparation of the whole research. It sets a number
of goals at the initial stages. Rather than a quick process of determining
research focus and questions, it is important to get the right focus and to
write well-designed questions through reflection and interaction.

Phase 2: Act to Implement the Plan

This phase involves enactment of the plans made in the previous phase.
To this end, Burns (2010) describes the systematic data collection pro-
cess and highlights the role of creating insightful links between data
collection methods and the focus under investigation. Regarding data
collection, regular classroom activities can be combined with the data
collection process in order to collect a natural set of data and so that stu-
dents also benefit while providing their responses in the form of views,
perceptions, practices, and knowledge. Specific data collection tools such
as observation (with different roles as an observer), brief notes of recorded
­comments during the instruction, audio or video recording of interaction
between students and teachers, maps, layouts, sociographs, or even pho-
tographs can be used. Other data collection tools are interviews (includ-
ing structured, semi-structured or open interviews), questionnaires and
surveys, as well as journals and logs and classroom documents.

Phase 3: Observe the Effects of Action in the Context

This phase is also known as data analysis and synthesis where action
researchers need to make sense of the information gathered through vari-
 1  Introduction to Action Research 
   5

ous data collection tools. Burns (2010) emphasizes, however, that AR

does not necessarily proceed in a linear order: it is often a cyclical or
spiral process in which data examination may start right from the very
beginning. Both qualitative and quantitative approaches to analysing and
synthetizing data may be used.

Phase 4: Reflect

The final phase involves reflecting on the whole process of research from
the very beginning to make informed interpretations of the data regard-
ing the issue under investigation. Burns (2010) emphasizes the impor-
tance of deep reflection in order to develop personal practical knowledge
and she makes a clear distinction between different elements in reflection
such as reflecting on practice, the research process, beliefs, values, feel-
ings, and experiences. These coherent layers of reflection are thought to
promote action researchers’ understanding of their classroom practices
and their own thinking processes.
Burns (2010) stresses that doing AR provides a natural cycle in that
it is systematically designed to develop practices on the basis of evidence
analysed and synthesized from the classroom-based data. From an AR
point of view, rather than a linear stage-by-stage development, each
cycle is interrelated with the previous or upcoming cycles. This also gives
action researchers the message that doing AR is a systematic and ongo-
ing process of refining ideas and practices. Action researchers, therefore,
develop awareness of different pedagogic elements that could be benefi-
cial to the process of evidence-based understandings. AR is therefore seen
as an iterative circle which shows how AR commences and develops in
a cyclical and dynamic manner. The iteration gives action researchers an
opportunity to revisit and rethink the process of researching and making
explicit links to their practices and beliefs by observing the experienced
influence.
Although Burns (2010) suggests a very clear methodology for doing
AR with specific cycles, there are still points that action researchers might
need to consider. This includes considering what AR is and what the key
characteristics of such research are. What follows is a succinct discussion
of some issues that could help researchers.
6  Developing Language Teacher Autonomy through Action Research

Planning AR
Although in the context of a real classroom environment a research proj-
ect does not always proceed according to pre-set stages, a well-thought-­
out plan can help researchers complete their research with fewer conflicts,
which means that they are able to focus on the content of the research
rather than having to deal with unexpected difficulties that may arise.
When planning AR, there are a number of issues to consider, depending
on the needs of a specific group, the particular context, and the charac-
teristics of the individuals involved. However, a teacher might need to
understand what AR is and how it differs from academic research.

Distinguishing AR from Academic Research

AR is different from research conducted by academics for general use in

the relevant academic area. The difference is made clear through the fol-
lowing comparative statements:

• AR problematizes a specific classroom issue, while academic research

problematizes a general one.
• AR aims to develop issues to promote quality of teaching and under-
standing, whereas academic research focuses on reaching generalizable
results.
• AR is seen primarily as self-informative, while academic research aims
to inform others too.
• AR requires a final reflection part where personal and professional
gains are discussed, but academic research requires a recommendation
part which discusses how findings could be useful for others.
• AR allows for personal pronouns, while academic research usually
requires the use of impersonal language and tone.
• AR aims to develop effective ways of teaching, while academic research
aims to prove what works well and looks for generalizable results to
wider contexts.
• AR focuses on contextualized issues, while academic research deals
with issues that concern all.
 1  Introduction to Action Research 
   7

• AR is generally conducted for professional development of classroom

practitioners, while academic researchers try to generate knowledge for
the relevant field of inquiry.

AR seems to be different from academic research in a variety of ways:

commencement, context, purpose, process, impact, and dissemination.

Commencement

AR starts with a classroom issue or a question which concerns teach-

ers’ own practices and understandings. It is usually initiated by the
teacher or by the request of the institution to develop the school qual-
ity through teachers’ own development. On the other hand, academic
research is conducted to fill the gaps in knowledge in the field, and it
may not be a teacher who does academic research. AR is usually initi-
ated as a bottom-­up activity by teachers who are willing to work for
better understanding and innovative activities that could foster help-
ful classroom practices. However, academic research could be initiated
upon the requests of supervisors, institutions, educators, or others who
are not participants in the research environment. Before commencing
an AR study, it is best to think about the questions and puzzles and
points to consider in the classroom. Starting research with issues gener-
ated form the actual environment will facilitate the investigation pro-
cess, help the teacher/researcher to maintain motivation, and promote
the quality of the findings.

Context

Another difference is that AR is conducted in classrooms where teach-

ers are teaching because they investigate a contextualized question, the
answers to which can be generated within the same context. However,
as academic research investigates issues that concern all, the context of
research may not be only one classroom but many. It may not be even
classrooms but a widely dispersed population which participates in the
research.
8  Developing Language Teacher Autonomy through Action Research

Purpose

The primary purpose of AR is to contribute to professional develop-

ment through developing awareness, knowledge, and practice. AR does
not seek generalizable results but promotes critical thinking skills. On
the other hand, academic research is expected to contribute to general
knowledge in a specific field with well-established tools and robust data
analysis processes.

Process

AR offers a particular process of doing research starting with identify-

ing the focus of research through critical reflection on experiences. In
AR, the process is thought to help teachers develop perspectives and
insights into the issue under investigation. It is a process-oriented
inquiry rather than product-oriented. The actual results may not be
as important as the cognitive and constructive experiences that the
teacher encounters during the process of engagement. According to
Dikilitaş (reported in Wyatt et  al. 2016), the benefits from teacher
engagement come “not only from the results of such research but from
the process of posing questions, hypothesising, looking for evidence,
synthesising all these, seeking for ways of innovating with existing
beliefs, knowledge and practice” (p. 7). On the other hand, academic
research employs a procedure which is well-planned and imple-
mented. Though researchers develop professionally when engaged in
the process of research, the ultimate aim is the results drawn from the
research. Its contribution to public knowledge is of critical impor-
tance, while AR, in essence, is a research activity that contributes
personal knowledge and growth.

Impact

AR leads to potential development of teachers, students, and schools

through teachers who collaborate with their students on a particular
 1  Introduction to Action Research 
   9

research topic. AR influences beliefs, knowledge, and practices of teach-

ers, which in turn contributes to student learning and development.
What is empowering and/or rewarding in AR is the potential for person-
alized and contextualized impact and direct involvement of the teach-
ers and learners in the learning process. On the other hand, academic
research may not consider teachers or learners to be primary beneficiaries
of their research though indirectly implications include them. Teachers
and learners may not even be able to access such research easily or under-
stand the language used. While AR may have a great deal of impact on
teachers’ and learners’ development, academic research may not be so
effective in creating direct impact. Such research needs to be mediated by
others such as trainers and educators.

Dissemination

AR is a form of research conducted by teachers with or without col-

laboration with mentors. Although some teachers may conduct AR
purely for their own information, others want to share what they find,
so it is important that teachers be given opportunities to share their
research with colleagues, educators, or students. The form of dissemi-
nation may be different from that of academic research studies. AR
could be disseminated through interactive posters, in online and web
conferences, or through face-to-face interviews, in addition to the
more traditional conference papers or journal articles. While making
research public is a must for academic research if it is to contribute
to the field, it may not be a priority for AR to be shared. However,
verbal or written dissemination of AR could also be beneficial to the
development of teachers.

Need for a Formal Course

There are diverse views on whether a formal course should be offered

before teachers start doing AR. The logic behind favouring formal train-
ing is that individuals should be equipped with knowledge about how to
10  Developing Language Teacher Autonomy through Action Research

do AR. On the other hand, a pre-AR engagement training may not be

useful as teachers may need to practise and see how theoretical knowl-
edge works in the real world. For example, learning how to collect data
and opting for particular data collection tools may not be realistic unless
they go to their classrooms and practise them. Only after such classroom
implementation can they combine theory and practice. It is this aspect
that has been highlighted in research carried out for professional develop-
ment. This is not to say that there is no need for a formal course, but it
should relate to practice in each stage.

Content

There is also a need for insightful discussion about the content of such
a course on AR. It is clear that it should encompass what it is, in what
specific cycle it works, what methodology it involves (including data col-
lection and analysis), and what interpretive methods can be followed. In
addition to these traditional aspects, a course content preferably contains
the aspects of exploration of the context, which will be discussed in detail
in Chap. 4 as well as the ways of writing it up for publication and dissem-
inating it for a wider audience, perhaps through innovative infographics
and online or conference posters which involve texts, visuals, and videos.
These aspects will be discussed in detail in Chap. 9.

Organization

An AR engagement may also involve thinking about the organization of the

process. A simultaneous progression is one of the options including learn-
ing about AR, doing it, and writing up the relevant section. For example, if
teachers are learning about how to delineate the context through collabora-
tive exploration of context and teaching, then they can be guided to record
as they explore by means of taking notes during the discussions with oth-
ers, which could be followed by writing it up before they engage in other
points of the cycle. There are also examples of learning and doing research,
which is followed by a writing-up process at the end of the research, and
sometimes presentation of the research at a conference.
 1  Introduction to Action Research 
   11

Length of Engagement

The length of AR engagement is also key to the benefits of doing research.

This may include growth in awareness and understanding of practical
aspects as a result of pilot classroom experimentation. Longer engage-
ment in a research focus may lead to more and deeper reflection over
a particular issue. Therefore, a full academic year could be a good way
of creating opportunities for insightful elaboration and informed reflec-
tion over topics of research, which may also allow for further consulta-
tion with others who may be of help. Equally importantly, the quality of
the process of engagement experienced should be considered. Bringing
together different views, perspectives, and knowledge through construc-
tive dialogues and creating a critical stance could be key to benefitting
from the longer engagement in the research process.

Sustainability

Doing AR is an engagement that requires critical thinking and active

teachers. The main positive aspect of AR is that it is contextualized in the
researcher’s school. As teachers may lack research competence, they may
need a longer period to complete an AR project. However, longer engage-
ment is a significant aspect of such research in that it gives the opportu-
nity to gain deeper insights into the research focus. Similarly, engagement
in AR in successive years is also a factor that contributes to develop-
ment as a teacher. Doing AR only once may not make good researchers.
Therefore, teachers could be encouraged to engage in AR for a longer
period of time in order to cultivate meaningful professional development
that contributes to their classroom practices, pedagogic understandings,
and depth of knowledge about teaching.

Critical Issues

There are also critical issues that an action researcher needs to prioritize.
Teachers doing AR might have a purpose for doing it so there is need at
some stages for stepping back and thinking about how much the research
12  Developing Language Teacher Autonomy through Action Research

progression addresses the expectations and initial purposes. Collaboration

with others constitutes a critical part of pedagogy, so teachers need to seek
ways of working with others throughout the research. As discussed in the
organization part above, generating relevant written work as the research
progresses could be useful in that the writing process could help formal-
ize emerging knowledge as a result of engagement in AR. Sharing with
others the initial and medial outcomes or findings of research is another
element that could relate to the quality of research, since constructive
feedback could enrich the content and facilitate the process of research.

Internal and External Support for Action Researchers

There are several sources of support for those who do AR, including an
internally hired trainer, an external supervisor, or experienced teachers
who have done several AR studies before. Though each can be of great
help in their own capacities, there might be limitations in the support to
be offered by an external supervisor who will probably be away from the
actual context most of the time particularly when teachers need imme-
diate access. Similarly, an experienced action researcher may be a good
research practitioner but may not have the necessary training qualities
and qualifications to guide other action researchers. However, an inter-
nally hired teacher trainer specialized in facilitating and supporting action
researchers could contribute to teachers’ engagement.

Collaborative AR
Collaborative action research (CAR) is an interactive form of AR as
opposed to individual engagement. While doing AR could be an individ-
ual endeavour, it could also be carried out with more than one researcher
who shares similar problems or questions. Focusing and elaborating on
shared issues constitutes a point where those in the community of prac-
tice can come together to distinguish contextual issues from individual
perspectives, which contributes to deeper understanding. Burns (1999,
p. 214) argues that “collaborative research offers opportunities for infor-
mal individual thinking to be transposed into more systematic and col-
 1  Introduction to Action Research 
   13

lective problem-solving”. CAR is seen as a thought-provoking process

whereby individual thoughts are naturally amalgamated through sus-
tained exchange of knowledge, experiences, and expertise.
However, CAR is also discussed as a form of collaboration between
a mentor or a supervisor and a student. This type of collaboration sup-
ports beginning researchers through expert knowledge constructively
exchanged among parties, which facilitates the process of overcoming
initial challenges. Mitchell et al. (2009) see CAR as appropriate and rel-
evant to novices who might need critical support while they strive to
survive the challenges through a zone of proximal development (ZPD)
where they are scaffolded by another person who knows more. For them,
CAR functions as a network of support backed up by mentors.

Individual or Collaborative

Doing AR is a process carried out by a researcher with or without any

external support provided by a supervisor, though Borg (2015, p. 108)
characterizes research as an experience that is collaborative, interactive,
constructive, practical, and scaffolded, implying that support is naturally
required. So although AR can be an individual experience depending
on the nature of the issue being investigated and the research skills and
confidence of the researcher doing it, collaboration can add to the depth
and breadth of ideas, analysis, and interpretations. Figure  1.1 sets out
diagrammatically some of the possible conditions for collaboration
AR is also situated in a specific context, such as a school, a programme,
or a university, which creates a natural social environment for research.
Researchers working together in an institution are likely to share similar
challenges and concerns, which necessitates a kind of collaboration, in
which they either do a joint AR or become critical friends for deeper
reflection. A key question which needs to be addressed is “Who, exactly
will be involved in this research?” There are advantages to working on
one’s own: increased control over exactly what gets done, not having to
“chase” people who “drag the chain”, and so on. But, of course, there are
also disadvantages. Research can be a lonely job, and it is often surprising
how a bit of brainstorming can generate answers that one would never
have thought of if working alone. Like everything, there is no right or-­
14  Developing Language Teacher Autonomy through Action Research

A common queson
A common interest
A common commitment
A common focus
A common purpose
Common expectaons

Common challenges in pracce

Respect for others’ opinions

Open-mindedness
Willingness to share and co-operate

Fig. 1.1  Possible conditions for collaboration

wrong answer to this, but, for those considering working in collabora-

tion, Fig. 1.1 might help with thinking through some of the important
considerations.

Potential Forms of Collaboration

There are several ways of forming collaborative groups among teachers,

students, parents, administrative staff, or others. The nature of collabora-
tion depends on the focus of the AR. One of the reasons for incorporat-
ing different parties into the research is their relevance to the research
and the degree of their contribution as a source of information, exper-
tise, or elaboration. Another reason could be that different parties share
similar concerns and they come together to investigate the issue in ques-
tion. Each party might have a different role and degree of contribution
to the research which creates a synergy for a deeper and multi-perspective
understanding of the aspects of the research. Burns and Hood (1997)
suggest three ways of working collaboratively:

• Research pairs: two researchers investigate a topic of interest for both.

• Research groups: individual or paired researchers come together to
exchange ideas and share perspectives.
 1  Introduction to Action Research 
   15

• Research teams: researchers from the same context gather for working
on a selected topic.

There are also other ways of creating collaborative groups in different

contexts. For example, collaborating with learner(s), which is one of the
key aspects of exploratory practice (EP) (Allwright and Hanks 2009),
takes an altruistic perspective. Research carried out with learners usually
takes the form of using them as subject and object of the data collection
process which may not benefit them at all. However, learner-integrated
research in which they engage in pedagogic tasks could also strengthen
the methodological robustness which adds to the quality of the data col-
lected. EP in this sense provides an organic setting for working together
with learners because exploratory practitioners utilize pedagogic activities
that they normally do in the classroom as central tools for generating data
to analyse. Learners as active participants of research provide an insider
perspective to the issue being investigated.
Alternatively, collaboration can occur between different institutions
such as schools and universities through setting up networks where teach-
ers can work on preparing projects, developing curricula, implement-
ing continuous professional development, and setting up face-to-face
or online interaction channels. Collaborative research projects can be
developed and sustained in order to bring together the diversity of ideas
and perspectives to generate empowering networks from which all parties
benefit.

Benefits of Collaboration

In general, collaboration offers opportunities for new learning, new

shared perspectives, and refreshed thinking and engagement in mean-
ingful partnership (Livingston and Shiach 2013). In this sense, it paves
the way for original ideas and perspectives provided by other minds for
a healthy co-construction of new knowledge. It is a form of bringing
together knowledge, experience, and expertise (Eady 1997, cited in Burns
2015). Collaboration may also add to the diversity of views in research
and shared workload. Though sometimes it could be hard for the collabo-
16  Developing Language Teacher Autonomy through Action Research

rators to synchronize with one another, the benefits can outnumber the
disadvantages. Teachers could be encouraged to work together especially
on shared classroom issues, which contributes to the multi-perspective
dimension of ideas generated.

Dissemination
Writing and presenting is often regarded as a natural follow-up to research,
since researchers may feel that what they have found needs to be commu-
nicated with others. The process of dissemination can be seen as a form
of reflection that is conducted after the actual research process. Preparing
a presentation and structuring the research process while addressing a
group of people requires careful planning whereby researchers deepen
their reflection by seeking ways of expressing it. Learning as a teacher
continues in the quest of refining the emerging pedagogic ideas formed
through the research experience. In addition, dialogue emerging during a
presentation leads to a natural context in which ideas are further refined
and clarified. If there are chances of question asking and answering after
the presentation, this could help the teacher by taking a third-eye view
into the intellectual process of research.
Learning through research engagement also develops as the teacher is
engaged in writing up the research as a potential form of written reflec-
tion. Writing is also considered an empowering process and tool of learn-
ing. Writing leads to reading, thinking, and putting into words through
synthesizing the ideas in the mind, and evaluating the way information is
transferred into written forms. Research experiences including how it was
done, what emerging knowledge it generated for the teacher are formally
expressed in a genre other than verbal.
How and when the act of writing up the research occurs could vary.
Some prefer to produce a post-research product, while others choose
to write as they complete specific parts of the research, which could be
described as simultaneous writing and researching. There is no particu-
lar answer to which strategy might lead to better learning. However, an
integrated approach which involves sustained research engagement and
reflection for writing up may seem to some a good option. These deci-
 1  Introduction to Action Research 
   17

sions relating to writing up research may vary depending on the context

and conditions as well as purpose of AR. Leading names in research for
professional development highlight the crucial role of writing up and
disseminating for further development not only in their classrooms and
schools but also in their overall career.

Quality of Research
Although there are debates about the quality of teacher research (TR),
the spread of research by teachers still continues. Those who advocate AR
by teachers usually consider the intellectual benefits gained during the
process and the wealth of practical knowledge developed by the teacher-­
researcher. Doing research is a skill that develops as one engages more and
more. Doing AR once will not make a good researcher. Creating quality,
then, is a matter of doing research with more commitment over time.
If we are to produce quality research, there are two essential concepts
which must be considered: validity and reliability. As Griffiths (2013,
pp. 197–198) points out, validity refers to the extent to which research
measures what it is supposed to be measuring. There are several different
types of validity:

• Construct/concept validity—how well the target concept is measured

(e.g. does a learning style questionnaire really measure learning styles?).
• Content validity—the extent to which the content is appropriate for
the context (e.g. a questionnaire designed for English as a Second
Language (ESL) students might not be suitable for English as a Foreign
Language (EFL) students).
• Convergent validity—the extent to which results accord with the
results of other studies (e.g. if I find a study done in New Zealand
found a relationship between anxiety and poor course results, can I
assume that this will also be true for my Turkish students?).
• Concurrent validity—the extent to which a study accords with others
done at the same time (e.g. is there a relationship between a study I do
with my class, and the same study done by another teacher with his/
her class at the same time?).
18  Developing Language Teacher Autonomy through Action Research

• Criterion validity—the relationship between the results of a study and

a given standard (e.g. an end-of-course or a standardized exam).
• External validity—the extent to which the results are generalizable to
a wider population (e.g. if I find that my Turkish students think it is
impolite to use a phone in class, is this also true of students in other
contexts?).
• Face validity—the extent to which a study appears to be measuring
what it is supposed to be measuring (e.g. a study which used the results
of a grammar test to assess speaking ability would lack face validity).
• Internal validity—the extent to which results can be used to infer cause
and effect (e.g. if I find there is a relationship between language learn-
ing strategies and end-of-course scores, can I conclude that good strat-
egies lead to successful course results?).
• Predictive validity—the extent to which the results of a study are able
to predict future outcomes (e.g. if I find that a group of students show
high levels of motivation, can I predict that these will be successful?).

As for reliability, this refers to the consistency of the data. There are two
main types:

• Internal reliability—this refers to the extent to which results are con-

sistent with each other. This is typically calculated by a test known as
Chronbach’s Alpha (see Data analysis, Chap. 6).
• External reliability—this refers to the extent to which the same con-
clusions would be reached by another researcher examining the same
data (sometimes also called “inter-rater reliability”).

There are those who do not advocate teachers’ engagement in research

because they claim teachers do not have the knowledge, skills, or time
required for doing research, so the research teachers do may not have
the desired rigour in terms of methodology. The criticism by those who
argue this way is that data collection and analysis carried out without
much awareness of rigour in research methodology could lead teachers to
generate misleading knowledge or implications, which could have nega-
tive effects on learning and teaching. This undermines the ultimate aims
of research by teachers, which is to develop learning and teaching quality.
 1  Introduction to Action Research 
   19

So these people see quality in research as a precondition to quality

in teaching and learning because AR constitutes a natural link between
teaching practices and the research process. What is critical to the pro-
cess of learning is therefore not just the research findings but the active
engagement in understanding classroom issues, which leads to poten-
tially insightful further reflections.
Borg (2010) suggests some criteria for a quality of academic research,
which are also required by AR or other forms of research. The criteria are
represented by the following questions to be considered by researchers.

• Is the purpose of the research clear?

• Have research methods been appropriately chosen and justified?
• Have data been collected and analysed in a technically competent
manner?
• Does the work make some kind of contribution to knowledge, with
potential implications for practice?
• Is the research ethical?
• Does the researcher adopt a critical stance?
• Is the research reported in a manner that is coherent? (p. 12)

Any research undertaken by professional researchers (teachers are no

exception) needs to address these criteria explicitly and clearly if they
are to produce a quality of research that yields reliable results that might
inform them. Following these criteria can also help teachers gain poten-
tial benefits more effectively.

Stages of AR
Although it should not be seen as a straightjacket, a view of the stages by
which AR can proceed can be helpful as a means of keeping a study “on
track”. The stages might follow each other as per Fig. 1.2.
Though AR is conceived as operating in cycles by Burns (2010), in
order to create a practical procedure for teachers, we have designed
AR in stages. The time allowed for each stage will depend on the local
situation. For instance, if a teacher wishes to complete a study within
20  Developing Language Teacher Autonomy through Action Research

Stages Activities

Stage 1: 1. Exploring the context

Planning the 2. Specifying research questions
research design 3. Deciding the research paradigm
4. Reviewing the literature
5. Deciding on collection procedures
6. Preparing tools for data collection
7. Obtaining consent and dealing with other ethical
procedures

Stage 2: Doing 1. Collecting the data

the research 2. Sorting the data

Stage 3: 1. Carrying out analysis procedures

Analysing the (e.g. means, medians, correlations - quantitative)
data (e.g. inductive & deductive content analysis -
qualitative)
Stage 4:
Interpreting the 1. Answering the research questions with evidence from
results the data
2. Drawing out implications (for teachers, students,
administration, curriculum/syllabus designers, etc.)
3. Considering limitations (what would be done differently
or what could be changed to obtain different results)
4. Looking into the future (what questions, puzzles have
emerged for further research or how will the research
help future development as a teacher)

Fig. 1.2  Stages of AR

 1  Introduction to Action Research 
   21

one semester (typically about 16 weeks), each stage would be about

4  weeks long. Of course, in a real teaching situation, things rarely
work out quite so smoothly: there may be all kinds of interruptions
and unexpected occurrences to deal with. But if there is at least a plan,
it can act as a guide against which researchers can monitor progress
and make adjustments if time is slipping by and necessary actions
have not been taken.

Types and Forms of AR

In AR teachers are expected to problematize a question, challenge,
dilemma, issue, or critical point that is considered to be causing dysfunc-
tion or presenting a challenge. However, there are several different forms
of research in which teachers might engage in their classrooms, and they
are labelled differently in the literature.
One such grouping sees reflective practice (RP) as the umbrella term
for the following forms of research in that all forms of research require
reflection.
–– AR functions as a way of enhancing practice by improving the ratio-
nality of practice.
–– Exploratory practice (EP) functions as a way of understanding and
thus improving the quality of life of teachers and learners as opposed
to improving or changing practice.
–– Lesson study (LS) functions as a way of developing practices at the
curricular level.
The justification is that all these three forms of research require reflection
over the course of research engagement. On the other hand, Cochran-­
Smith and Lytle (1999, p. 22) see teacher research as the superordinate
term for AR, RP, and EP.
In this book, AR is seen as the umbrella term since it may include all
practitioners in different fields who would like to investigate their own
contexts to take actions and create deeper understandings. These are dis-
cussed in detail in the following sections.
22  Developing Language Teacher Autonomy through Action Research

Teacher Research (TR)

TR has been defined by different authors, each highlighting a different

aspect. Holliday (1994) considers researching one’s practices as being
critical to developing informed pedagogic decisions:

In order to arrive at appropriate methodologies, practitioners need to take

time to investigate what happens in the classroom. They need to incorpo-
rate into their approach the capacity to look in depth at the wider social
forces which influence behaviour between teachers and students. (p. 17)

Cochran-Smith and Lytle (1999) define TR as an umbrella term that

includes

all forms of practitioner inquiry that involve systematic, intentional, and

self-critical inquiry about one’s work in…educational settings. (p. 22)

Bailey (2001) implies that regardless of research methodology, any

research done by teachers for professional development to promote
teaching and learning practices can be considered TR:

Teacher-research has been used increasingly in recent years to refer to any

systematic investigations conducted by teachers, regardless of the method-
ological approach employed. (p. xiii)

Similarly, Borg (2013) refers to TR as

systematic inquiry, conducted by teachers, into some aspect of their own

context, with the aim of improving both understanding and practice.
(p. 201)

Borg (2013) discusses different conceptualizations of TR.  First, TR is

conceptualized as a tool that could promote social change whereby teach-
ers feel “liberated” through the freedom of researching any local personal
and contextualized issues that may not be touched upon by academic
research in the field. Second, TR is perceived as a way of knowing through
a collaborative form within a particular context such as classrooms or
 1  Introduction to Action Research 
   23

schools. According to this view, TR can help teachers promote practices

of teaching in their school contexts. Finally it is also seen as a vehicle for
teachers to revisit, reconsider, and explicate practical knowledge encapsu-
lated in their own existing practices.
According to Borg (2015), TR begins with a question or an issue that
teachers want to learn more about and understand better. The question
may emerge from a problem, although TR may not offer a solution to
it directly. Teachers, nevertheless, may want to understand more about
something that works well. The second stage of the cycle is to act by col-
lecting and analysing data for the question initially posed. Borg describes
TR as being empirical in that it is centrally driven by data collection
and analysis. The next stage is making sense out of the data collected
and seeking new insights which may help inform pedagogic decisions.
According to this model, Borg (2015) also sees TR as cyclical rather than
linear, which allows for sustained, iterative engagement in the issue.
Borg (2015) discusses reflecting, reading, communicating, and col-
laborating as additional processes that could support the cycle of TR. By
reflecting, Borg means systematic and continuous thinking about
research, while by reading he refers to reviewing what has previously
been said about the focus of the research, which could be regarded as
the literature review in academic research. He also sees communicating
as a process of talking to colleagues to get feedback on their own work
and inspire them to engage in their own inquiries, while also highlight-
ing the role of collaborating in the sense that teachers work together to
produce joint projects. Borg in his model prioritizes asking questions fol-
lowed by collecting and analysing data. After that, the teacher-researcher
is expected to make sense of the analysed data in order to take informed
action, which is the last stage.
There is no explicit focus on the writing up and dissemination processes
in either Burns’ cycle or Borg’s model. The writing up and presentation
are particular processes which could also be rewarding for the teacher.
Presenting research develops the confidence of teachers and provides
feedback from different people, which could promote their critical think-
ing and awareness. Similarly writing up allows the teacher-researcher to
revisit and reconsider procedures and interpretation of research, which is
also a process during which teachers learn and develop their understand-
24  Developing Language Teacher Autonomy through Action Research

ing of the topic under investigation. More specific positive impacts of

writing up on teachers is shown by McLean and Mohr (1999), who sug-
gest that writing up helps identify and articulate new understanding(s),
helps to improve classroom practice, describes new understandings that
emerge from the research, helps us to clarify our own knowledge, facili-
tates an investigation of exploration of the classroom implications, and
facilitates inter-collegial collaboration.
For those who want more detail about the writing-up process, see
Chap. 9.

Exploratory Practice

EP is a professional development tool that can be used in any context

ranging from education to medicine to law. It can be a tool for under-
standing more about an issue in collaboration with all parties in that
particular context with specific attention to what is not normally under-
stood without consulting others. Even though we sometimes know, for
example, reasons or causes of an issue, this does not necessarily mean that
we should not investigate it. In an educational context, Allwright and
Lenzuen (1997) define EP as a way of conducting:

classroom investigations which provide language teachers (and potentially

learners also) with a systematic framework within which to define areas of
language teaching that they wish to explore, to refine their thinking about
them and to investigate them further using classroom activities…as the
investigative tool. (p. 73)

The definition highlights the use of normal pedagogic activities, rather

than more traditional academic research techniques, as investigative
tools, which is thought to give the researcher (practitioner) an already
familiarized tool by which to investigate the issue. Furthermore, such
“normal” activities are likely to be more acceptable to learners, who may
resist more intrusive methods (such as random assignment to experimen-
tal groups or tasks which they do not see as relevant to their learning) for
which they do not see any immediate benefit to themselves in terms of
their own learning.
 1  Introduction to Action Research 
   25

Allwright (2005) highlights the function of EP as providing a basis for

further change. He suggests that striving to understand classroom life will
provide opportunities for teachers and learners to collaborate pleasantly
and productively. He also suggests that this interaction can prepare an
insightful basis for smooth pedagogic change, although one cannot make
changes at the exploration stage: the purpose of this stage is to prepare
one’s cognition for any instructional changes to be made in the future. In
addition, to call a pedagogic issue a problem without adequately explor-
ing and reflecting on it could be risky. The necessity of exploring and
reflecting before investigating the issue is necessary at this stage as well.
Allwright highlights this issue by suggesting a softer and less negative
word, puzzle, instead of problem.
A puzzle in the field of EP is referred to as an issue into which a prac-
titioner would like to gain a greater understanding through pedagogic
activities. Puzzles arise from classroom challenges, dilemmas, questions,
critical points, and pedagogic curiosities that teachers have had or their
experiences of teaching, which go unattended and lead to a need for
understanding. Puzzles may not be problems that lead to failure in learn-
ers and teachers, but areas to expand understanding to boost confidence.
Puzzles require “puzzling” which refers to reflecting on situations and
asking “why” questions about them, rather than rushing into looking
for “solutions” (Allwright and Hanks 2009, p. 176). Puzzles in this sense
are springboards from which teachers can make plans for understanding
(Allwright 2003).
EP offers a principled framework for teachers/practitioners who
would like to understand more about the classroom, learners, materials,
and themselves by puzzling about teaching experiences. The principles
include general characteristics that help teachers/practitioners under-
stand the underlying logic behind it.

• The primary principle is characterized by the term “quality of life” for

language teachers and learners. When teacher and/or learner puzzles
are elaborated, the understanding constructed and shared by both par-
ties may lead to depth and breadth in the insights into the puzzling
issues. Teachers and learners can raise more awareness into teaching
and learning issues that play a critical role in development.
26  Developing Language Teacher Autonomy through Action Research

• The second principle is prioritizing quality of life rather than degree of

success as measured by grades and scores. Quality of life is contrasted
with quantity of output in the sense that standard output-based learn-
ing may be more important than developing quality of life for teachers
and learners (Gieve and Miller 2006). Similarly, Allwright (2003)
argues against improvement-based education and says understandings
of life mean life is the big issue, not improvement.
• The third principle relates to developing collegiality and mutual devel-
opment, the need for which is emphasized by Allwright (2005). EP is
seen as a form of research which brings together academics and
­practitioners (teachers, learners, and researchers) to elaborate on prac-
tical issues arising from the teaching contexts (see Allwright 2003;
Hanks 2009). Collegiality benefits all these parties in terms of under-
standing one another and developing chances of learning.
• The fourth principle focuses on understanding the sustainability of
research and the embedded interrelation between the pedagogy and
the research. Since exploratory research activities are conducted in
class time, teaching naturally becomes researching. EP recommends
the use of already practised classroom activities as investigative tools.
Such embeddedness is methodologically robust as it provides a reliable
set of data which is collected from the natural context of research.
• The fifth principle is based on the practical aspects of planning research
using potentially exploitable pedagogic activities, or PEPAs (Allwright
and Hanks 2009, p. 157). PEPAs are critically developed to address
the central puzzle which forms the basis of the research. The principles
of EP are categorized as what, who, and how issues, (Hanks 2015),
which make the roles and focus points even clearer in terms of the
process of engagement in EP. The specific division is as follows:

The “what” issues

–– Focus on quality of life as the fundamental issue.
–– Work to understand it before thinking about solving problems.
The “who” issues
–– Involve practitioners developing their own understandings.
–– Work to bring people together in a common enterprise.
 1  Introduction to Action Research 
   27

–– Work cooperatively for mutual development.

The “how” issues
–– Make it a continuous enterprise.
–– Minimize the burden by integrating the work for understanding
into normal pedagogic practice.
• In addition to these principles, it is clear that teachers go through an
exploration process before they engage in research. Therefore, a sixth
principle can be added to this list, which is:

–– Seek critical areas which need researching further.

This principle is important in that it can help teachers do AR in previ-
ously under-researched areas that are critical to their classroom instruction,
although exploratory practitioners may opt for skipping some of these stages
if they are not relevant for their own particular needs and environment
Allwright (2000) identifies seven systematic stages for EP, the first of
which involves identifying a puzzle. Procedures that underlie EP, as would
be the case in the initial stages of any kind of research, involve a curious
desire to learn more about teaching and learning issues in the classroom.
These might include an attempt to understand the causes of

• low student motivation

• negative stances towards testing and assessment
• lack of commitment to the learning process
• dislike for particular activities handed out by the teacher

and so on. The list could be extended across several other specific issues
depending on the teachers’ own experiences of teaching and the class.
The curiosity to gain deeper insights could also include positive aspects of
the context such as why a particular activity worked so well or the reasons
for students’ dominant preference for a particular

• instructional method
• material
• activity
• interaction pattern.
28  Developing Language Teacher Autonomy through Action Research

A critical point to remember is that a puzzle does not necessarily have

to be motivated by a problem or an instructional weakness, which could
discourage practitioners. A puzzle is a pedagogic area about which teach-
ers would like to explore more, which might give them an opportunity
when planning their lessons based on their learners’ and their teachers’
expectations from the lesson.
The reflection stage can be characterized as “puzzling about” (Allwright
2000) so that practitioners can have a wider perspective before they take
an action. After a puzzle has been identified, the relevant issues can be
explored to provide a springboard for in-depth understanding. The reflec-
tion at this stage can be carried out through reflective dialogues with all
relevant parties including learners, teachers, coordinators, managers, and
perhaps parents. The reflection could also be done individually through
a series of observations of the classroom focused on learners or teachers,
through analysis of already produced formal and informal documents by
students, teachers, administrative staff, and parents’ written accounts (if
any). The sources of information might vary according to the puzzle and
the contexts practitioners work in. A carefully performed exploration and
reflection can contribute to the depth and breadth of the understanding.
Once exploration and reflection have been done (which could even
continue at this stage), practitioners can place special attention on the
specific issue by narrowing down the focus. This stage involves collect-
ing naturally occurring data about what is being puzzled about through
monitoring, during which extensive field notes can be kept. More specifi-
cally, practitioners identify data sources and ways of gathering data in the
next stage.
This stage involves reconsidering the degree of need for further data
and taking action if the researcher has not been able to understand
enough about the puzzle using the planned data generation procedures.
However, planning data collection is not understood as a process that
requires isolating it from normal classroom activities. Allwright (2000)
recommends trying to find classroom activities that will produce data
which can themselves be used for research purposes. This highlights the
role of EP as a form of research that does not involve the same intervening
methodological strategies for data collection, which could be positioning
learners as object and subject of conventional research (Hanks 2014). EP
 1  Introduction to Action Research 
   29

places greater emphasis on the role of learners in the data collection pro-
cess as co-researchers in the inquiry. To this end, EP suggests integration
of normal pedagogic exercises and activities which seem to be natural
enough to support normal teaching plans. Such a technique could save
time, produce reliable data from a naturally occurring environment, and
give learners confidence and autonomy by which they might feel respon-
sible and committed to the resolution of the puzzle.
Critical to the data collection process is self-reflection on the normal
pedagogic activities done during instruction, and finding or creating
PEPAs which learners like to engage in (see Table 1.1).
Choosing the most appropriate ones for EP will allow for a normally
occurring process and achieve deeper insights and reflection from learn-
ers. PEPAs might help generate useful data which will help to develop
ways of looking at learners’ understanding of the puzzle. Some examples
of potential PEPAs can be seen in Table 1.1.
It is also possible to help learners to identify their own puzzles and
increase their understanding of it, which could contribute to their quality
of classroom and learning life. Once students have a better understand-
ing of their dilemmas, they can feel more committed to learning and can
optimize their potential for motivation and learning. There is need for
developing natural ways of helping them to reflect upon their puzzles.

Table 1.1  Examples of PEPAs

Data from activity and
Verbal data Written data Data from task exercise
Pair work Diaries Games Tests
Group work Self-evaluation Projects True and false exercises
Role play Suggestion box Field trips Jigsaw reading
Consultancy Problem box Vlogs Sentence completion
Discussion Dialogue journal Blogging Gap-filling sentences
writing
Interviews Learner-to-learner Self-­recording Reading texts about
correspondence the puzzle
Storytelling Story completion Videotaping
Brainstorming Writing critical Drawing
incidents
Self-evaluation Peer feedback
Adapted from Allwright and Hanks (2009)
30  Developing Language Teacher Autonomy through Action Research

These might include tasks, activities, and exercises, whether verbal or

written, which they most like to engage in.
By this stage of the research, we need to consider what we have
achieved so far and decide what to do next. In conventional research, this
would be called the data analysis and interpretation stage. EP ­researchers
are expected to reflect on the data collected to decide whether it will
be deep enough to understand the puzzle. They will also assess whether
more data could be needed to promote the quality of understanding.
This stage seems to be a strategic one at which the practitioner needs to
reflect insightfully in order to reach conclusions and critical decisions. It
is equally important at this stage to collaborate with others in order to
foster the depth and breadth of understanding, which will provide an
intellectual basis for further critical elaboration.
By now, the practitioner may need to step back and assess the relevance
and potential of the conclusions arrived at. The degree of understanding
may be addressing the nature of the puzzle and improving the “quality
of life”. Meanwhile, the practitioner could also communicate the initial
results to learners and see how they are responding to the issue. The prac-
titioner needs to make insightful interpretation to understand how much
his or her initial expectations encapsulated in the identified puzzle have
been met as a result of action for understanding.
Though going public is often seen as the last stage of a research project,
as a result of sharing the results with others, it may also function as the
beginning of new puzzles or deeper understanding that may lead in dif-
ferent directions. There may be different motivations to go public. One is
that it may help practitioners develop ways of utilizing the understanding
in further pedagogic experiences. Another is that practitioners may feel
that the understanding of the puzzle is not enough, so the understanding
arrived at could be shared with others who could participate in the pro-
cess of greater and more adequate understanding. The modes of delivery
of the understanding vary according to the opportunities already existing
in the context. These might include a classroom project presentation to
which other teachers or learners are invited, preparation of an interactive
poster to share with other colleagues, attending a local or national confer-
ence to reach a wider audience, submitting to a journal, or planning an
online web sharing system to discuss the ideas with geographically diverse
practitioners.
 1  Introduction to Action Research 
   31

Exploratory Action Research

First developed by Richard Smith in a year-long project with Chilean

secondary school teachers (see Smith et  al. 2014), exploratory action
research (EAR) involves extensive initial discussion relating to practice.
It also encourages teachers to do research-based exploration of classroom
issues or puzzles (Allwright 2003) in ways which do not interfere with
normal classroom teaching processes, and only then to consider bring-
ing about and evaluating a change in practice. A similar description of
the role of exploratory research was also highlighted in Dikilitaş (2015)
with the idea that the issue in question is discussed and explored from a
broad perspective which forms a basis for a more specific focus for further
research.
Smith (2015) characterizes EAR as a gradualist approach to teacher
inquiry. The initial concern of EAR is to lead teachers to extensive explor-
ing and understanding before they think about and plan for how they
can change their existing classroom practices and observe and evaluate
the impact. The exploration stage, therefore, is critical in that it might
provide potential clues for further actions.
Wyatt and Pasamar Márquez (2015) interpret EAR as an eclectic com-
bination of EP and AR as EAR includes several aspects from both, but it
should not be considered simply as a mathematical addition of one to the
other, as indicated by Smith (2015), who describes it rather as a holistic
strategy emerging from contextual needs arising during the Chile project.
On the basis of this experience, Smith suggests EAR as a workplan that
could be integrated into inquiry-based teacher development projects in
other contexts where AR or EP on their own may not be applicable or
sufficient (see Rebolledo et al. 2016 for examples intended to encourage
other teachers).

Reflective Practice

RP is rooted in the educational insights inspired by John Dewey (1933,

1938), who sees the act of learning as a process of insightful reflection
based on a series of experiences. According to Dewey, thinking r­ eflectively
requires describing and questioning experiences systematically, which
32  Developing Language Teacher Autonomy through Action Research

then leads to construction of further hypotheses for which new actions

are created and implemented to see the results. Engagement in reflection
is closely related to the task of teaching where teachers usually need to
think about what and how they have taught in order to inform future
practices. This is seen as a continuous process in which teachers need to
engage to promote their teaching and in turn student learning.
Reflection is recognized as a tool in educational settings for foster-
ing teaching practices through insightful thinking about the practices
in the classroom. It helps teachers generate personal pedagogic and new
meanings synthesized with experiential knowledge. RP is interrelated
with engagement in AR in the sense that action researchers need to think
deeply about the dimensions and aspects of the research focus with ref-
erence to context, including students, other teachers, and institutional
priorities. Reflection is seen as the catalyst to initiate the problematizing
which is required for the process of doing AR.
RP seeks to provide more solid clues and a sound basis for the prob-
lem/puzzle to be investigated systematically before a teacher decides on
a pedagogic change in classroom practices. RP supports the ideas that
teachers can learn by closely reflecting on classroom experiences (Schön
1983; Wallace 1991). RP provides teachers with opportunities to gain
insight into their experiences to better understand their practices. Such
an activity can promote teachers’ understanding of the issue in question
from students’ perspectives as well as from their own subjective perspec-
tives, providing insights from multiple sources. The exploration through
discussion in EP as an initial stage is consolidated with more focused
and closer examination of a specific practice in one’s own teaching. The
emerging or identified weaknesses or points to improve at this stage can
better inform the problematized issue to be investigated in AR, which
can follow exploratory and RP research experience.
RP involves “profound commitment to the process of discovering and
analysing (Bright 1996), verbalising professional knowledge (Loughran
2002), and investigating a more insightful dimension than trial and error
approach” (Raelin 2002, p. 66). Teaching experiences are the basis from
which a process developed and enriched through reflection is initiated
to help teachers and students learn and develop. To this end, particular
steps are followed to initiate and sustain RP in teacher development. For
 1  Introduction to Action Research 
   33

example, Gibbs (1988) introduces the following stages inspired by Kolb’s

experiential learning cycle. He sees experience as the initial element that
activates the reflective process, which is followed by a stage of descrip-
tion of what happened without a judgmental stance. Then he suggests
describing personal reactions and feelings about the experience without
detailed analysis, followed by an evaluative approach to identifying good
and bad sides of the experience. Having set a detailed basis, the reflec-
tive practitioner can now start analysis, which involves making sense of
the experience through consulting others in the context. This is thought
to help understand what really happened and what others experience in
similar situations. This is followed by drawing general and specific conclu-
sions about the experiences and analysis. The final stage in RP is design-
ing personal future action plans on the basis of the lessons learnt from the
reflective process.
Engagement in systematic reflection on classroom experiences is a form
of professional development on the basis of the personal and contextual
act of thinking from multiple perspectives. Reflection-based development
contributes to the process of teacher learning through self-informative
understanding with an insightful formative evaluation of experiences.
It also helps teachers become independent of external impositions of
development learning. Initiating a reflective stance could lead to deeper
understanding, which is then synthesized with future teaching practices.
Teachers can make informed decisions about their own teaching through
the use of their own reflective capacities. From this perspective, teach-
ers become self-knowledge generators with minimized impact of expert
knowledge and develop autonomy by regulating their own development.

Lesson Study

LS is a form of reflection supported by peer observation to develop better

understanding of actual classroom settings at the moment of teaching
with the specific purpose of developing as a teacher. The major char-
acteristics include collaborative reflection and student centeredness. It
creates teacher groups who observe a lesson during teaching and focus
on ­developing lesson plans on the basis of the weaknesses and strengths
34  Developing Language Teacher Autonomy through Action Research

observed. This lesson is called a “research lesson” in that teachers observe

and reflect constructively on the lesson and think jointly about how the
lesson can be developed. The key difference from other forms of research-
based professional development—possibly a limitation—is that it is
focused on student learning only, whereas other forms of research such
as AR, RP, and EP consider all potential aspects which bring together the
teacher and student engagement in the process of investigation.
However, as reported in Cajkler et al. (2014), LS is also a way of devel-
oping opportunities for helping students overcome their difficulties, for
teachers to plan in detail, for in-depth reflection about the quality of
teaching and learning (Lewis 2009), and for building up a collaborative
environment for teachers to develop a sense of collective efficacy (Lewis
et al. 2006).
Engagement in LS provides indirect benefits for teachers in that the
major focus is on observation of students to enhance student learning.
As opposed to teacher-oriented investigation, LS prioritizes students and
their learning process. The general practice of LS also involves post-lesson
interviews with students to collect their ideas which are not observable by
the teacher observers.
Having discussed different approaches to doing research, Table  1.2
might help us see all these forms of research for professional development
and compare the cycles and/or stages involved in the operationalization
of each.

Table 1.2  Forms of AR

Forms of
research Cycles and/or stages
Action Plan Act Observe Reflect
research
Teacher Ask Collect Analyse Make sense Take action
Research questions data data
Exploratory Identify the Design Collect Analyse data Develop
Practice puzzle PEPAs data understanding
Reflective Experience Describe Analyse Draw Design action
Practice conclusions plans
Lesson Formulate Plan Observe Reflect
Study goals for
curriculum
 1  Introduction to Action Research 
   35

Teacher Autonomy
Teacher autonomy is generally defined as teachers having control over
their own professional development and practice, especially in terms of
developing independence and interdependence in their own context (e.g.
classroom), when they need to make autonomous decisions for how they
plan to teach, what they practise in the classroom while teaching, and
how they can improve their teaching practices. Teacher autonomy has
been defined in different ways:

• Little (1995) sees teacher autonomy as a teacher’s capacity to engage in

self-directed teaching.
• Benson (2000) sees it as teachers being able to keep themselves away
from control that could be imposed by others.
• Smith (2003) describes it as a concept that relates to teachers’ capacity
to engage in self-directed teacher learning.

According to Smith (2003), critical reflection provides teachers with

opportunities to foster autonomy particularly for developing metacog-
nitive skills about how they can learn as a teacher. Using investigative
tools to elicit learners’ views of learning, for example, could be one of
the major ways of ensuring autonomy for teachers. Such teacher learning
engagement is critical because it is both reflective and understanding-­
oriented. It is this kind of learning that might enable teachers to access
direct learner-related data. AR, in this sense, provides ample investigative,
discovery-based, and inquiry-based learning experiences that promote
teacher autonomy.
There are several teacher development activities that have the poten-
tial to lead to autonomy depending on the depth and breadth of and
commitment to professional engagement. Doing teacher-led research is
one of those. For example, engagement in AR is an empowering way
of creating freedom, being self-directed, and becoming learner-teachers
because teachers might be able to generate their own practical knowledge
through exploring their own contexts, reflecting on their practices, and
questioning their classroom teaching. There seems to be a close relation-
ship between AR and teacher autonomy, in that AR engagement has the
36  Developing Language Teacher Autonomy through Action Research

potential to help teachers build such self-confidence to deal with critical

questions and problematic issues in the classroom through going deeper
into the issue on the basis of data collection and analysis either by the
teacher or by a joint investigation with learners.
How teachers develop autonomy is an issue recently discussed and
investigated from multiple perspectives. Autonomy can be gained either
during the engagement in research or after the engagement in research.
Different dimensions of autonomy can emerge through engagement in
AR.  One can develop autonomy to identify the focus of and purpose
of research with decreasing support from a mentor and critical friend.
Another area could be to develop autonomy in developing a research
plan, operationalizing it with data collection tools, and analysing the sets
of data. Another could be to promote autonomy in making pedagogic
connections between what findings say and how these may inform the
classroom practices of the teacher-researchers. We see autonomy not as
one single dimension but as having multiple sub-dimensions for each
step of research conducted, ranging from identifying the research focus
and purpose to drawing relevant implications for teaching.
It is, however, important to remember that autonomy does not mean
that a teacher can do as he/she likes without considering the consequences
of actions on students, colleagues, and other stakeholders. In fact, a teacher
is in a highly responsible position, where his/her actions have the ability to
affect others in quite profound ways. It is essential that this responsibility
is always taken very seriously, and that teachers always take great care to
remain accountable to those who might be affected by their actions.
The relationship between engagement in AR and teacher autonomy is
not explicitly and empirically shown in the literature. In fact, it is almost
impossible to measure levels and degrees of autonomy that could be exerted
by research engagement. However, a positive mutual relationship is estab-
lished between the engagement in AR and developing teacher autonomy
(Wang and Zhang 2014; Lieberman 1995). Such c­ onnection is enabled by
the positive impact of engagement in AR on the teacher. Doing AR is a way
for promoting inquiry and reflection in order to develop understanding,
investigate and experiment with practices in the classroom, which helps
teachers make critical practical changes both for themselves and for their
students (Burns 2015). In this way, AR engagement helps teachers under-
 1  Introduction to Action Research 
   37

stand themselves better as teachers. The insights gained through AR experi-

ence create valuable opportunities for teachers to develop skills in adjusting
their instructional decisions and material selection and use. Autonomy
manifests itself in different ways as teachers perform their profession with
more and more agency and ownership in pedagogic issues. Lieberman
(1995) argues explicitly for the relationship between teachers actively ini-
tiating and carrying out AR in their own schools and classrooms and the
potential to promote teacher autonomy.
AR allows for teachers doing critical reflections upon their teach-
ing practices and emerging research results. Teacher autonomy as well
as learner autonomy can be potentially developed through a reflective
approach to teacher development once the link between teacher devel-
opment and autonomy is explicitly established. Smith (2000) describes
teacher autonomy as becoming aware of and accessing sources of learning
through critical reflection and investigative activities (Vieira et al. 2008).
Vieira et al. (2008) sees teacher autonomy as an outcome that can be
achieved through engagement in critical reflection which can facilitate
liberation and empowerment of teachers with enhanced interest and abil-
ity to overcome contextual constraints in the teaching–learning context.
In this sense, teacher autonomy is potentially linked to the reported ben-
efits of AR, which also involves a high degree of reflection for teacher
development.
More recently, Wang and Zhang (2014) report how collaborative
AR—a highly participatory, interactive, cooperative and educational
process—constituted a major source of developing teacher autonomy for
student teachers. Wang and Zhang described the way to become more
autonomous with three stages.

• Stage 1 involves student teachers feeling excited with their own

puzzles.
• Stage 2 describes the difficulties encountered during AR and the per-
sistence exerted by them.
• Stage 3 is the achievement stage where student teachers gain experi-
ence and self-confidence with expanding insights as a result of con-
ducting collaborative AR and develop new inquiries with growing
enthusiasm.
38  Developing Language Teacher Autonomy through Action Research

Within this framework, Wang and Zhang (2014) describe gradual

autonomy development with active engagement in doing AR.
McGrath (2000) conceptualizes teacher autonomy as self-directed
professional development and argues that such professional development
could be achieved by the following teacher roles:

• teacher as researcher (Stenhouse 1975)

• teacher as action researcher (Burns 1999)
• teacher as reflective practitioner (Schön 1983, 1988).

McGrath establishes direct links between autonomy development and

the role played by teacher-researchers. In addition, Carr and Kemmis
(1988) suggest that Stenhouse’s concept of teacher-researcher in Britain
affirmed and justified a well-developed sense of professional autonomy
and responsibility.
Vieira et al. (2008) argue with reference to Parker (1997, p. 39) that
AR engagement allows for theory generation rather than theory applica-
tion, a perspective that requires teachers to be autonomous rather than
dependent on others’ theories and apply them in their classrooms. In this
way, AR provides potential avenues for developing autonomy through
investigation and experimentation to generate knowledge. Vieira et  al.
(2008) highlight AR as one of the powerful ways of promoting teacher
and learner autonomy.
Therefore, teachers’ ability to promote their teaching practices in the
classroom could be closely related to the developmental and professional
activities in which they have the roles of knowledge seekers and builders
through AR. Engagement in AR can boost confidence and capacity to
reflect and investigate practical knowledge whereby to build a personal
theoretical knowledge.
This book also argues that autonomy can potentially be developed when
teachers engage in AR by carrying out systematic inquiries about their own
practices. The degree of autonomy to be gained from engagement could be
enhanced by increased rigour in the methodology of AR. In the book we
document various ways of engaging in research and promoting research
skills whereby methodologically robust AR can be carried out.
 1  Introduction to Action Research 
   39

Though AR engagement contributes to the development of ­autonomy

in teachers, this does not necessarily mean that AR is the builder of auton-
omy but rather the booster of it. Engaging in AR, teachers can pave the
way for possessing more autonomy in their professional development as
well as in their teaching practices.
AR, then, promotes teachers’ professional autonomy in the following
ways:
–– combining independent and interdependent decisions about criti-
cal aspects of teaching
–– enabling practical knowledge strengthened by critical reflection
upon practical knowledge
–– inducing significant evidence from AR that could inform practical
knowledge and teaching practices
–– promoting abilities to access sources of different views and opinions
through collaborative discussion with all parties involved in the
teaching context
–– strengthening critical insights into what learners want, how they
learn and develop
–– cultivating intellectual skills to link practical knowledge to theoreti-
cal knowledge or vice versa
–– sustaining knowledge growth and practical development across pro-
fessional careers

Feasibility
When it comes to deciding the various issues related to AR, a key
but often neglected issue is that of feasibility (can it be done?).
Constraints may include the size of the class, the location of the room,
teaching load, timetable clashes, institutional requirements, parental
expectations, student resistance, examination deadlines, or any of the
multiple other demands on a teacher’s time and energy, including per-
sonal constraints such as family or ongoing study. In fact, it is no use
designing a wonderful piece of research which is simply impossible
40  Developing Language Teacher Autonomy through Action Research

in a particular environment or circumstances. So, when beginning to

think about a research project, it is necessary to consider the practical
constraints there may be.
However, with proper planning constraints that might inhibit engage-
ment in doing research for professional development can be overcome.
To enhance feasibility, it is possible to collaborate with colleagues and
administrative staff to find joint solutions to the inhibitive factors that
might interfere. In addition, collaboration with an experienced mentor
or researcher can help with AR and suggest how to do it in an effective
way. Alternatively, specific guidelines that explicitly show how to proceed
to do an AR especially in the initial years of engagement can be helpful.
It is known that teachers can develop research self-efficacy beliefs in time
with sustained active engagement and a gradually developing sense of
agency and ownership.

Sustainability
There are a number of studies that report the positive impact and
practical and cognitive changes which AR engagement might lead
to in teachers. However, what is equally important is how to sus-
tain changes and developments through an ongoing approach.
Sustainability is a concept used to highlight the role of sustaining
research endeavour and understanding more about practices, beliefs,
and knowledge over years. From a different perspective, Edwards
and Burns (2016) refer to “sustainability” as sustaining the positive
impact achieved from the participation in AR programmes even after
they are completed. However, Edwards and Burns (2016, p. 14) also
discuss sustainability as “identifying sustained impacts at organiza-
tional, national, and possibly international levels”. They also argue for
a balance of bottom-up individual teacher motivation and top-­down
institutional support, which might contribute to the degree of the
sustainability of the impact of AR over time.
Yuan and Lee (2015) suggest the need for the integration and
sustainability of positive impacts of AR by the school’s continuous
 1  Introduction to Action Research 
   41

professional development programme. According to Edwards and

Burns (2016), such sustainability can be achieved by the following
considerations:

• sustained application of AR framework for confidence building

• ongoing use of communication encouraged by AR
• recognition of teachers AR globally
• opportunities for writing up and publication
• school-based sharing of AR studies

Sustainability relates to long-term effects of AR on teachers, which could

be interrelated to the context in which teachers work. Sustainability of
these potential impacts in schools is to be ensured through resourceful
facilities including incentives, promotion, support, and recognition of
work and granting autonomy and responsibilities to the teachers who
engage in AR.  To this end, teachers with AR engagement experience
need to be supported in several ways so that they keep integrating ben-
efits of AR into their work in the classroom and share their work with
colleagues nearby in different forms (see Chap. 9 for opportunities for
dissemination).
Similarly, Allwright (1997, p.  369, as cited in Edwards and Burns
2016) argues for the development of a “research perspective” by which
he implies that teachers need to look at their practical issues from a
researchable point of view, which he also describes as “ongoing concerns
for undertaking”. There are few studies that report on the extent to
which action researchers sustain positive impacts they gained through
doing research. Sustainability can be ensured through considering the
following:

• establishing strong links between the new knowledge and the practice
• linking the insights into skills which are transferable to different
situations
• identifying potential areas where the new knowledge might be applied
• highlighting the benefits from reflection
• creating opportunities for reflection, discussion, and research
42  Developing Language Teacher Autonomy through Action Research

• seeking collaborative opportunities

• taking positive impacts further

Sustainability implies long-term impacts of what has been learnt or

achieved. Sustainability that can be generated through AR will require
sustaining the improvement and modification of practices and develop-
ment of changing beliefs and knowledge. However, sustainability might
also mean the need to promote new understandings as the positive
impacts need to be changed in the future. AR will provide a number of
areas to think about, which can help continued development as a teacher.

Benefits of Sustainability

Dikilitaş (2014) reports considerable sustained engagement in TR with

support focusing on developing positive attitudes towards research,
developing research skills and critical thinking, and heightening teachers’
awareness of their learners (Dikilitaş 2015; Smith 2014). Sustainability
seems to contribute to deepening and developing positive benefits of
research engagement. By establishing interconnectedness between what
has been learnt from AR engagement and what is done in the class-
room, there are more chances of sustainability, of ongoing engagement
in research, and successful integration of positive benefits into teaching
practices.

How Sustainability Can Be Achieved

Dikilitaş and Mumford (2016) discuss how the opportunity for teachers
to present and publish their research helped sustain their engagement in
doing AR. Setting up goals for teachers such as deadlines for writing up
for publication and holding conferences at particular times of the year
could be indicators of motives behind sustainability. There are several
ways of supporting sustainability. Some include

1. making use of positive psychological and educational benefits of

research
2. engaging in AR systematically
 1  Introduction to Action Research 
   43

3 . presenting and publishing AR for recognition and autonomy

4. recognition of AR by the institution
5. building a professional community of AR
6. providing ongoing dialogues among action researchers and

colleagues

These six forms of support can help sustain AR engagement and the
positive pedagogic benefits which derive from it. It is clear that engage-
ment in AR is a multidimensional activity that involves not only teachers
but also learners, institutions, and other people. Ensuring sustainability,
therefore, depends on

• personal interest and motivation: motivation and interest is needed in

sustaining professional development over a longer period of time. A
sense of career-long learning is required, as otherwise teachers can just
repeat themselves without much development in teaching and think-
ing abilities.
• sustained dialogues with colleagues: continuing interaction with col-
leagues could also help to ensure sustainability in professional develop-
ment. Formal and informal talks with other teachers in the institution
might lead to shared understandings that could be empowering.
• relevant practical talks and collaboration with learners: AR engage-
ment can also be sustained through developing talks with learners
about the specific issues of the AR.  Continued implementation of
outcomes is required particularly to see how and to what extent they
work. Follow-­up implementation can provide evidence about modi-
fied instruction. Collaboration with learners might lead to deeper
insights into the understanding of the questions, puzzles, or issues
from the perspectives of learners.
• quality of ongoing support from the institution: institutions are integral
contributors to teachers’ professional development. By recognizing
teachers’ AR results and showing willingness to consider them in cur-
riculum design, institutions provide fundamental support in enabling
sustainability. Teachers’ sustained engagement in AR which is also sup-
ported by institutions further contributes to the developmental cul-
ture of institutions.
44  Developing Language Teacher Autonomy through Action Research

Questions to Consider
1. Can you think of some issues in your professional environment that
you might problematize?
2. What are some of the factors you will need to consider when planning
your research?
3. Will you want to collaborate? If so, how, where, when, with whom?
4. Will you want to disseminate your findings? If so, how, where will you
do this?
5. How will you protect the quality of your research?
6. How long do you think each of the stages in Fig. 4 might take you?
7. Which type/s of research is/are most appealing to you in your profes-
sional context?
8.  How do you think conducting research might contribute to your
autonomy?
9. To what extent do you think your autonomy could overlap with your
responsibility and accountability?
10. How will you deal with feasibility issues given your professional con-
text and the issue you want to research?
11. What do you think you can do to ensure sustainability in your teaching
context?
 2
Clarifying the Research Questions
or Hypotheses

Research is a systematic process of understanding questions growing in

the minds of researchers. It is the research question that triggers one to do
research. Selecting or identifying research questions is the initial stage of
developing a research plan. It is a critical process because the chosen topic
plays a role in completing research. If researchers problematize issues they
are curious to learn more about, they are more likely to sustain motivation
to do research. The impact of doing research will also be maximized as they
encounter answers they are looking for throughout the process of researching.
An important source of research questions for teachers is the teaching
experiences they go through. Access to teaching experiences is a great
opportunity for action-researchers. These experiences may arouse curios-
ity and interest in teachers about particular research questions.

 hinking About Questions, Hypotheses,

T
and Points to Develop
It is not uncommon for research to begin before researchers have a clear
idea of what they want to find out. In fact, deciding the research topic

© The Author(s) 2017 45

K. Dikilitaş, C. Griffiths, Developing Language Teacher Autonomy
through Action Research, DOI 10.1007/978-3-319-50739-2_2
46  Developing Language Teacher Autonomy through Action Research

and clarifying the question/s or hypothesis/es can often be the hardest

part of any research project. But clarifying the research question/s or
hypothesis/es early in the process can save a lot of wasted time, prevent
missed opportunities, and avoid overlooking important details which
may cause problems later on.
What is the difference between a research question and a hypothesis?

• A research question is exactly what it says: it asks a question and is

punctuated with a question mark. A research project requires at least
one question, but there may be several (Nunan 1992).
• A hypothesis contains the researcher’s prediction/s (Dörnyei 2007).
According to a null hypothesis, there will be no difference between the
groups being researched. If a difference is found, the null hypothesis
will be rejected, and the alternative hypothesis will be supported.

In fact, hypotheses are more common with experimental studies (see

Chap. 3). However, experimental studies can be difficult to set up in
real teaching contexts, since students are typically already in established
classes, making random assignment difficult, and experiments depend
on some kind of intervention which may interrupt normal classroom
procedures and strike resistance from students, parents, or school
authorities. Also, there may be ethical issues (see Chap. 5) regarding
consent and/or differential treatment of students, especially if some
students are perceived to be given advantages that are withheld from
others. Action research, therefore, is much more commonly (although
not exclusively) descriptive in its nature, that is, it aims to describe a
phenomenon, perhaps by means of questionnaires, or it may take a
qualitative approach and use observations, interviews, or case studies
(see Chap. 3). This kind of research is much more likely to start with a
question or questions to which the researcher wishes to find an answer,
rather than trying to ­predict what the researcher thinks the answer is
likely to be. For this reason, this chapter will deal mainly with the issue
of research questions.
It is also important to discuss the distinction between theoretical ques-
tions and practical questions. The former refers to questions that are
answered to develop new theories or test existing theories and to discover
 2  Clarifying the Research Questions or Hypotheses 
   47

generalizations (Ary et  al. 2010). Such goals could be too challenging
particularly for language action-researchers. Some examples of theoretical
questions include:

• What is motivation?
• How does learning occur?
• Why do young learners forget easily?

On the other hand, practical questions deal with the actual problems
encountered at the level of practice in education (Ary et al. 2010), which
fits into the purpose of research conducted by teachers. Such questions,
when answered, could help teachers or educators find evidence for under-
standing issues, problems, and puzzles. The answers could also promote
teachers’ or educators’ knowledge and understandings of pedagogic
issues, which provide them with opportunities to develop their teaching.
Some questions of this kind are:

• How beneficial is brainstorming before product-based writing?

• What is the effectiveness of video-enhanced materials in teaching
speaking?
• What is the role of teachers in direct feedback in learning
pronunciation?

These questions are quite relevant to the practical aspects of education

that might contribute to teachers’ development.
Action-researchers need to make such distinctions at the ques-
tion formulating stage, as theoretical questions may not be properly
answered with the data sets collected and the research design estab-
lished. Teachers, on the other hand, are not usually expected to deal
with theoretical issues but with practical issues which they might need
to theorize personally.
This chapter will help teachers to get started in the initial stages and to
find a focus point. It aims to promote teachers’ skills to ask appropriate
research questions with a wide range of sample questions. It will also pres-
ent some guidelines as to how to ask “good” research questions.
48  Developing Language Teacher Autonomy through Action Research

When Should Research Questions Be Written?

Some say that writing research questions is the beginning of research,

while others say research questions can be written at any stage as data
collection and analysis proceed. Though research questions function as a
guide as one does research, sometimes one realizes that the data set is not
going to answer the research questions, which requires revising or even
changing them. If the research is a qualitative one based on interviews,
observation, or other forms of verbal or written data, then writing up
research questions may be subject to change as themes from the data
start to emerge. Instead of having research questions, the purpose of the
research can be drafted, which could be more helpful at the initial stages.
However, in quantitative research studies, which involve pre-set surveys
and tests, identifying research questions at the beginning could really
help with the research. The questions may hardly change (apart from
wording) as the data collection tools shape the data that will come out.
For example, a question such as:

• To what extent do students improve their writing skills during the

term?

could have a set answer framework such as low, medium, or high improve-
ment: the answers to the question will only vary along this continuum.
With a qualitative research question, however, the answers are almost
never known (except that the researchers may have some assumptions).
The emerging pattern from the responses could help to shape the research
questions around the purpose of the research.
Overall, research questions may be revised and changed in a qualitative
study, while in quantitative research it is usually more possible to set the
questions at the beginning and investigate accordingly.

What Is the Research for?

Becoming aware of what is being investigated is a critical stage where it is

necessary to rethink and identify what is being explored by investigating
 2  Clarifying the Research Questions or Hypotheses 
   49

it. This helps to shape the research questions. Here are some ideas to
think about:

• What are students’ opinions, views, attitudes, or perceptions regard-

ing particular issues of concern to the teacher and/or the other
students?
• What are students’ reactions to the activities they are asked to do?
• Have there been/should there be any changes and developments in
teaching practices?
• Have there been/should there be any changes and developments in
student learning?
• Have there been/should there be any changes and developments in
students’ learning processes?
• What is the impact of specific practices on students’ knowledge, skills,
understanding, or learning?

More can be added to this list of issues to be investigated, allowing focus

to remain on particular issues.

Types of Research Questions

Researchers might create research questions that address several different

dimensions. It is important that researchers are aware of these dimen-
sions. In this way they can also develop awareness of what they are
investigating.

Questions for Understanding

Common in exploratory practice, questions for understanding aim

to develop researchers’ knowledge about context including learners,
materials, pedagogic activities, or even colleagues working in the same
­environment. This could be the initial attempt to develop understanding
of relevant issues by teachers. Questions might seem very general, leaving
room for further and deeper thinking. For example:
50  Developing Language Teacher Autonomy through Action Research

• What do my students think about my classroom management?

This could be a question that could create curiosity in teachers’ minds

and help them understand classroom management from students’ per-
spectives. Similar questions can be generated in a way to include col-
leagues such as:

• What do colleagues think about the course books we are using?

• What do administrative staff think about the students’ development?

Such questions could be a strong basis for further research that investi-
gates issues, looks for evidence, and provides a basis for further actions
and decisions in the future.

Impact Questions

Research questions might involve teachers, learners, and colleagues, as

well as activities, tasks, techniques, methods, or materials. These two sets
of issues are actually inseparable. Rather than focusing on each, research-
ers could find a way of combining both in research questions. Questions
that relate to impact are good examples of this combination. When
researchers look into the impact of their practices on learners, they seem
to manage to investigate the issue from two sources: materials and people.
This is very much aligned with the notions of teachers doing research for
professional development in that such research allows them to create and
investigate connections between what they do and how learners benefit
from this. Teachers might want to investigate the impact of what they do
in the classroom. For example:

• What is the impact of audiovisuals in the beginning of the lesson on

learners’ reading comprehension?

The investigation could be based on self-reported data sets from the stu-
dents but also on the grades they receive from the follow-up quizzes or
tests. In this way researchers might compare the findings from students’
self-reports and score-based data sets.
 2  Clarifying the Research Questions or Hypotheses 
   51

Such questions help teachers develop their practices and understand-

ings about their own practices. They could find evidence that enables
them to improve their teaching and develop their mindset. A higher
impact coming from scores or grades could mean using audiovisuals,
while self-­reports could provide detailed perspectives to consider, includ-
ing challenges experienced and benefits gained.

Perception Questions

Looking into perceptions is another way of forming research questions.

Understanding perceptions could help teachers learn more about their
students’ preferences. Negative or positive perceptions towards teaching
practices and teachers’ behaviours in the classroom could bring to light
several implications for teachers to consider. For example:

• What are the students’ perceptions towards working in pairs or groups?

This could help teachers understand how students perceive pair work or
group work, by which they can plan and adjust their classroom interac-
tion in the future. Similar questions can be developed on the basis of
perceptions of students, teachers and administrative staff, or parents. The
collected perceptual data might lead teachers to be more self-critical and
inform them of the changes and developments they can make in their
professional practice.

Improvement Questions

Improvement-based questions can be generated by teachers. Some

researchers might want to know how students improved as a result of
direct exposure to their practices. These questions might involve using
interrogatives such as how much, how many, to what extent, or how
often and usually enable teachers to learn about learners’ progress. For
example, researchers might want to compare two groups of learners
who were exposed to two different types of instructions offered in dif-
ferent classrooms. They compare the scores of different assessment tools
52  Developing Language Teacher Autonomy through Action Research

during the terms and calculate the numerical improvement. A question

that investigates this focus could be:

• Which vocabulary instruction type led to higher grades among stu-

dents in different classes: integrated or isolated?

Such questions usually lack depth and breadth because they do not
answer why such a result was obtained. However, investigating improve-
ment is essentially quantitative, whereas looking into learner develop-
ment might involve more qualitative data generated through interviews,
diaries, documents, and so on.

Identifying Questions

Teachers might also be interested to explore some aspects of their con-

texts by asking open-ended and non-leading questions. These questions
are open to any perspective that could emerge from the data set. Teachers
do not set off with assumptions. For example:

• What are the factors that affect students’ motivation?

This could be a good identifying question which naturally creates a curi-

osity in the researcher rather than trying to prove issues they already had
in mind. Closed-ended questions imply a particular type of answer in
one way or another, whereas such open-ended questions make researchers
wonder about what will emerge. Another such question could be:

• What insights can we gain into students’ vocabulary learning abilities?

With this question researchers might want to explore the issue in more
depth.
Instead of asking “what are the students’ vocabulary learning strate-
gies?”, which looks into only the strategies students use, we can leave
room for more perspectives such as factors, advantages, and challenges in
learning vocabulary with a “what insights question” aimed more generally
 2  Clarifying the Research Questions or Hypotheses 
   53

at ability. Such questions with a potential to explore in more depth allow

for better understanding of relevant pedagogical issues for teachers.

Purpose and Research Questions

Research questions are sometimes mistaken for the statement of purpose.

The purpose concerns itself with the general focus and ultimate goal of the
research, namely, the major reasons for which the research is conducted.
On the other hand, a research question is a way of narrowing down the
purpose of research into researchable items which are then dismantled
to address and discuss the purpose. From this perspective, research ques-
tions are those that bring together the overall purpose of the research. To
shed light on this issue, the following example could be given.
The purpose: What can I do to develop my students’ writing skills?
Research Questions:

1 . What teaching practices do I conduct in my writing lessons?

2. What is my students’ level of motivation for writing?
3. What kind of writing activities do they like to engage in?
4. What alternative teaching practices can I plan and conduct in the
lesson?

Answering all these questions might provide a number of ideas about the
purpose (developing students’ writing skills). The first question allows
reflection and self-criticism, where the second one helps understanding
of the kind of support needed to provide for them. The third one helps
understanding of what students want in writing lessons and would pro-
vide information to plan accordingly to maximize the opportunities for
them to benefit from the teaching, whilst the fourth one urges improve-
ment in teaching practices by creating new ways to teach writing.

Delineating a Research Question

In a recent online teacher research training supported by Teaching English to

Speakers of Other Languages (TESOL), there was an opportunity to analyse
questions generated by participants. One of the questions was as follows:
54  Developing Language Teacher Autonomy through Action Research

• How can I increase my students’ participation in speaking lessons?

The question seems to be the general purpose of the research because this
is the ultimate aim the researcher had in mind. To address and attain the
purpose, there is need for more focused questions.
It also sounds quite summative, which means it looks into how before
understanding what actually happens in the classroom, what teachers do
and what students think about it. Without understanding these aspects,
it is not very effective to think about “how” the situation can be devel-
oped or changed. Therefore, some sub-questions could help the teacher
understand the phenomenon.
As part of detailing the questions for better exploration and reflection,
the following questions can be asked:

1 . What activities do I (as teacher) perform in speaking lessons?

2. What are the students’ preferred activities in speaking lessons?
3. What are the students’ behaviours in speaking lessons?
4. What activities do they like and dislike in speaking lessons?
5. How do inter-student relationships affect willingness to participate in
speaking lessons?
6. What are the major emerging issues that I need to consider?

After having answered and understood the answers to these questions,

answers can be developed to the main question initially asked (How can
I increase my students’ participation in speaking lessons?). The ultimate
purpose could then be achieved by synthesizing all the answers to the
research questions.
By breaking down the question into small pieces that could be explic-
itly answered, the action researcher can now bring them together to make
insightful interpretations that could help understand the “how”, and
think about ways in which participation in the speaking lesson can be
increased.

1. Consider the contrasting opinions, needs, and expectations and make

strategic decisions to come up with alternative ways of increasing
participation.
 2  Clarifying the Research Questions or Hypotheses 
   55

2. Think about whether it is the teaching methodology or the activities

used that discourage students from participating.
3. Think about whether it is students’ lack of confidence, knowledge
and/or experience, or students’ relationships with each other which is
causing the lack of participation.

The insights gained through such a process of asking concrete and answer-
able questions could deepen perspectives and encourage investigating
the issue not only from students’ perspectives but also from the teacher’s
point of view. Understanding the role of the teacher in the process of
what is not working or needs to be developed will also create a demo-
cratic setting where educational practices are mutually developed by the
critical contribution of learners as well.

Research Question Guidelines

It is actually not as easy to write research questions as one might initially

suppose. “Good” research questions need to be clear, unambiguous, and
answerable. Some guidelines to consider:

1. Make sure the question is worth asking. To take a rather extreme

example for the sake of making the point, it may be that a class has an
unusual number of people with unusually coloured eyes, and, if this
variable is correlated with their exam scores, it is possible that there
might be a statistical relationship. But would this be at all a useful
thing to know? And would it be generalizable beyond this particular
group of students? It is important to consider such questions in the
interests of research which is actually worth doing.
2. Ask one question at a time. For instance, for the question “How can I
encourage my students to read and write more?” it is possible to find
the answer to one of these puzzles, but not the other. It is better to
frame the query as two separate questions.
3. Avoid negative questions, which can be difficult to answer from the
student’s point of view, and difficult to interpret from the researcher’s
perspective. For example, if we are trying to find answers to the
56  Developing Language Teacher Autonomy through Action Research

question “Why do students not participate in speaking lessons?”, and

a student responds to the questionnaire statement “I do not believe I
am a good speaker” by saying “No”, does it mean that this is not what
he/she believes, or that he/she agrees that he/she is not a good speaker?
Better to phrase the item positively with something like “I believe I am
a poor speaker.”
4. Make sure the question can be answered. As noted in Chap. 1, feasi-
bility is a key consideration when designing a research project. For
instance, if we are interested in finding the relationship of age to effec-
tive phonological development, it is probably not a useful question if
we are working in a university preparation school environment, where
the vast majority of students are likely to be no older than in their
early 20s, and even if there are a few older students, there will proba-
bly be too few of them to support reliable generalizations.

Questions to Consider
Look at the following research questions taken from actual teacher research
studies and consider the following:
(a) Are they good questions?
(b) If not, how could they be improved?
(c) Would they be appropriate for your teaching situation?
(d) How could you reformulate them for your context? Choose three of
them that interest you.
Sample research questions generated by actual practising teachers who
did action research.
1. How effective is data-driven learning on grammar teaching?
2. Is there a change in the students before and after the experience of
creating their own videos?
3. How effective are modern and traditional approaches to teaching
phrasal verbs on students’ learning?
4. Which corrective feedback method makes my students less annoyed?
5.  Does my students’ reflecting on motivation help them become
motivated?
6. What are the factors that cause my students to have difficulty in speak-
ing? What can we do to help them overcome these factors?
 2  Clarifying the Research Questions or Hypotheses 
   57

7. What kind of impact does peer assessment have on my students’ pre-

sentation skills?
8. What are the causes of my students’ reluctance to take part in group
work activities? How could I design effective group work activities?
9. How does team teaching influence teaching?
10. What are the constraints that prevent students from becoming autono-
mous? How can I promote learner autonomy?
11. What are the ways to improve teaching pronunciation through using
error correction?
12. What do students think about my using implicit and explicit ways of
teaching vocabulary?

Possible Questions (to stimulate ideas):

1. Why are some students unmotivated?

2. How can my students be encouraged to read more?
3. What is the effect if they do read more?
4. Everyone complains that our students don’t talk enough. What can
be done to encourage them to speak?
5. How can students’ vocabulary be expanded?
6. My students often make grammar mistakes in their productive lan-
guage. How can their accuracy be improved?
7. My students don’t like writing. What can I do to encourage them to
write?
8. Do students learn best through implicit or explicit teaching
methods?
9. How do my students’ learning styles relate to their achievement?
10. My students find it difficult to listen to native speakers. What can I
do to help them?
11. Some of my students do not have very good learning strategies. What
can I do about this?
12. Some of my students display signs of anxiety, and I feel that this
holds them back. What can I do to make them less anxious?
13. How can I encourage my students to be more autonomous, since I
cannot be there all of the time for all of them?
58  Developing Language Teacher Autonomy through Action Research

14. What gender differences are there, and how does it affect their
learning?
15. What is the effect of technology (e.g. their mobile phones or social
media) on my students’ learning?
16. How do relationships within the class affect classroom dynamics?

Practice Task 1
Develop research questions for the following purpose. Pay attention to
the potential links between the questions.
How can I develop learners’ autonomy in my classroom?
Relevant Research Questions:

1. .
2. .
3. .

Practice Task 2

Literature about language learning strategies says adults are more skil-
ful in developing language learning strategies than young learners. What
hypotheses or predictions can be developed for this?
Relevant Hypotheses or Predictions

1. .
2. .
3. .
4. .

Example: An Action Research Study on Classroom Culture—

Research Questions
A study conducted in a university classroom may help to clarify by exempli-
fication the ideas presented in this book. This example is not intended as
demonstrating the “right” way to do action research, because there really
is no such thing. But, since a picture is worth a 1000 words, it is hoped that
this example might help to illustrate and clarify some of the ideas
presented.
 2  Clarifying the Research Questions or Hypotheses 
   59

Background
The study was conducted among university students, firstly with a third-
year class (which was treated as the pilot) and later with a fourth-year class.
The issue
A classroom is a specific sociocultural environment which has its own set of
rules, language, expected behaviour, and standards of politeness. If these
standards are not met, it may stop us from working well with others whose
norms regarding what is polite or appropriate are different from our own.
There was a group of students in the third-year class whose behaviour
was causing concern, both to the teacher and to the other more serious
students. These problem students often came late, if at all, spent a lot of
class time fiddling with their phones, talking, and apparently paying little
attention to what was going on in the class. In response to concerns raised
by some students who found this behaviour distracting, the teacher initi-
ated a discussion on the standards of behaviour expected, and gathered
some of the ideas presented. These ideas were then written into a question-
naire, which was then administered to the class, followed by discussion.
The following semester, a fourth-year class was required to carry out a
small research study as part of their course requirement. For many of the
students, this was the first time they had done such a task, so they were
anxious and uncertain how to go about it. In order to provide a model to
guide them, the teacher decided to use the earlier study as a pilot, made
some adjustments to the questionnaire, and administered it to the fourth-
year class.
Research Questions
The study was basically concerned with exploring student perceptions of
appropriate behaviour. However, the issue was also raised about possibly
different perceptions by males or females. Furthermore, not all of the stu-
dents were from the country where the study was taking place, so differ-
ences according to nationality were also a possibility. Since gender and
nationality are easy to obtain, these were added to the questionnaire form.
In addition, in order to add a qualitative dimension, students were asked to
comment on their reasons for their opinions, and these comments were
then examined for insights which they might add to the quantitative rat-
ings data. The research questions for this study, therefore, were:

• What are the expected standards of behaviour in a university

classroom?
• Do these standards vary according to gender?
• Do these standards vary according to nationality?
• What reasons do students give for their opinions about appropriate
classroom behaviour?
Further details about this study will be presented progressively through-
out the book in the appropriate sections.
60  Developing Language Teacher Autonomy through Action Research

Task
Using your own teaching context, identify a problem/puzzle/issue/question
that you have to which you would like to find a solution or answer.
 3
Research Methods: Options and Issues

There are numerous different research methods, and each of them has
advantages and disadvantages. Although it may not be necessary to know
about or use all of them, a general knowledge about the options that are
available will help make informed choices about which one is most suit-
able for the context and research purpose. With this in mind, we will
outline the most commonly used methods here and provide an example
of each one so that they can be checked for their suitabilıty for a given
purpose.

Quantitative Versus Qualitative Research

Qualitative and quantitative research are two contrasting paradigms to
be followed in research engagement. While in a quantitative paradigm,
research questions are answered objectively by providing numbers, includ-
ing percentages, ranks, or means; in a qualitative paradigm the questions
are answered through careful analysis of what people say in response to
open-ended questions. These two paradigms differ in purpose, design,

© The Author(s) 2017 61

K. Dikilitaş, C. Griffiths, Developing Language Teacher Autonomy
through Action Research, DOI 10.1007/978-3-319-50739-2_3
62  Developing Language Teacher Autonomy through Action Research

Table 3.1  Comparison of quantitative and qualitative research paradigms

Quantitative Qualitative
Purpose To study relationships, cause, To examine a phenomenon as it
and effect is, in rich detail
Design Developed prior to study Flexible, evolves during study
Approach Deductive; tests theory Inductive; may generate theory
Tools Uses preselected instruments The researcher is a primary data
collection tool
Sample Uses large samples Uses small samples
Analysis Statistical analysis of numeric Narrative description and
data interpretation

approach, tool, sample, and analysis. Table 3.1, adapted from Ary et al.
(2010, p. 25), displays these differences.
Though quantitative and qualitative research designs contrast in
many ways, this does not necessarily mean one is a superior paradigm
over the other. Rather, they complement each other. For example,
quantitative research provides us with results as levels, degrees, quan-
tities, frequencies, percentages, ranks, and so on, whereas qualitative
research supplies results in the form of verbal statements concerning
feelings, ideas, reactions, impact, opinions, attitudes, and so on. For
example, imagine we found the percentages of students who are highly
motivated or un-motivated in the classroom. How will this result help
us understand the causes of low motivation and factors that affect
high motivation? At this stage we need qualitative data results, which
could give more information about contextual factors that led to such
motivational levels in the classroom. Without understanding these
underlying causes and factors, the numerical results will be insuffi-
cient to understand the phenomenon under investigation. So rather
than showing one of the paradigms as being superior to the other, it
could be wiser to think that both have strengths and weaknesses and
to make use of the potential of each in our research. The purpose and
research questions will also determine which paradigm we need to
follow or at what point we need to resort to both.
Quantitative research: An example of a quantitative study is the first
section of Griffiths (2016) about strategies for developing writing
skills.
 3  Research Methods: Options and Issues 
   63

The study reported in this article took place in a key university in China
where 59 sophomore (second-year) students studying in a writing class
were surveyed by means of a custom-designed Likert-­type questionnaire
regarding the strategies they used for developing writing skills in English.
The questionnaire was administered to the students during class time and
the ratings were analysed for medians (appropriate for nonparametric,
ordinal data such as are produced by Likert-type questionnaires) and also
correlated (using Spearman’s rho for nonparametric data) with end-of-
course scores. The results indicated that a small number of strategies were
significantly positively correlated with successful course results (in other
words, they were used by the more successful students), and another strat-
egy group was significantly negatively correlated (that is, they were used by
the less successful students).

Qualitative Research

Qualitative studies provide a holistic picture of the issue under inves-

tigation and allow for deeper understanding of the factors, insights,
roles, qualities, impact, contextual constraints, and advantages.
Teachers who conduct action research might need to collect qualita-
tive data from their students to get a fuller picture of their thoughts,
preferences, and perspectives in order to inform their future instruc-
tion. Qualitative studies aim to explore events, processes, and activi-
ties from the participants’ point/s of view. A typical question in such
research would be How did you feel about the use of reading strategies
in the class? The students are the participants and the researcher is the
teacher, who aims to understand his/her use of reading strategies as
experienced by the students. Key characteristics of qualitative studies
include:

• a wide range of qualitative data collection techniques, such as observa-

tion, interviews, think-aloud protocols, case studies, narratives, eth-
nography, or examination of documents (e.g. students’ written work
or journals—see later sections of this chapter).
• coding and looking for recurring themes in the data sets.
• categorization of emerging patterns and themes.
64  Developing Language Teacher Autonomy through Action Research

Research suggestions could include studies which focus on how learners

perceive their experiences in the classroom. For example, teachers could
collect data from their students using different data collection techniques
to understand how they perceive their classroom management, inter-
action patterns, instructional language, vocabulary teaching, or other
teaching practices.
The second section of the study by Griffiths (2016) on writing strate-
gies described above is qualitative.

After the quantitative analysis stage of the writing strategy study described
above, four of the top-scoring students were asked for their comments on
the questionnaire results. These comments were analysed for salient themes
which were then categorized according to whether they were positive,
neutral or negative. Although the responses from these four top-scoring
students were generally in the same direction as the quantitative results,
there was considerable variation in terms of individual strategy prefer-
ences, with complex inter-individual and dynamic intra-individual patterns
being apparent. The article concludes by suggesting the need to acknowl-
edge the complexities of strategy deployment, to respect individual strat-
egy variation, and to allow for dynamic strategy development in the
classroom.

Cross-Sectional Versus Longitudinal

Cross-sectional methods look at the relationships between/among variables
at a particular point in time. An example of a cross-sectional study is
Griffiths and Incecay (2015) about how styles and style stretching are
related to successful learning.

This study attempted to address the question of the relationship between

learning style and successful learning outcomes. It was conducted among
106 Turkish university students, using an original questionnaire constructed
using elements from established instruments. According to the results,
three styles were significantly correlated with exam results. In addition,
higher-scoring students reported a more eclectic range of styles, suggesting
more willingness to style-stretch, while lower-scoring students reported a
more limited range, a difference which also proved to be significant.
 3  Research Methods: Options and Issues 
   65

Longitudinal methods examine particular phenomena over time.

An example of a longitudinal study is Griffiths (2015) about strategy
development and progress in language learning.

This article reports on a study conducted in a private English language

school for international students in Auckland, New Zealand, which
aimed to investigate how progress in language learning related to
changes in reported frequency of language learning strategy use over a
period of time. Using a questionnaire known as the English Language
Learning Strategy Inventory or ELLSI, students were surveyed on entry to
the school. Thirty students were surveyed again three months later. It
was found that those who made the fastest progress through the levels
of the school were those who most increased their strategy use over the
period involved.

Experimental Versus Quasi-Experimental

The most traditional form of research is the experiment. This type of study
has very particular requirements, including:

• random assignment (participants are randomly assigned to particular

groups)
• using control (which receive the standard form of instruction) or
experimental groups
• applying some kind of intervention to the experimental group in
order to determine whether the treatment results in different
outcomes
• using a pre-test and post-test in order to assess the effect of the inter-
vention. In some cases, a delayed post-test is also added to see if the
effects of the treatment are sustained

An example of an experimental study is Yu (2009) about the use of lexical

phrases.
66  Developing Language Teacher Autonomy through Action Research

This study investigated whether two different means of learning the

word sequence “despite the fact (that)” led to different outcomes for
the acquisition of the phrase. Following a pre-test, the participants were
randomly divided into two groups. One group was taught by means of
explicit instruction, while the other was asked to commit the word string
to memory through repetition or recitation without receiving explicit
instruction. Following the post-­test, the recitation group was found to
significantly outperform the instruction group in terms of procedural
knowledge whereas the instruction group outperformed the recitation
group on declarative knowledge. The results suggest that memorizing
whole phrases may facilitate acquisition of the imbedded structure on a
procedural level, whereas explicit instruction can only promote declara-
tive knowledge without necessarily facilitating the ability to use the
knowledge.

Quasi-Experiment  It can be difficult to set up true experimental studies

since in actual teaching/learning situations students are usually already
assigned to classes, and providing different treatments for students
within a class can be problematic. This makes random assignment dif-
ficult. As a result, it is more common to work with intact classes, and
when this happens, the term quasi-experimental is used. Typically, either
the teacher will be teaching two classes at the same level and may decide
to use one as the control and the other as the experimental group, or
the teacher may compare results of his/her own class with those of a col-
league’s class which was taught according to standard methods. Using
two different teachers, however, introduces a confounding variable—
any effects may be due as much to the teacher as to the treatment, so
it is actually better to use one teacher if at all possible. Another prob-
lem may be that students may resist being used for research purposes,
although this resistance may be avoided with careful explanation (e.g.
the teacher wants to try out a particular method which may help them
to learn more effectively).

An example of a quasi-experiment is Tammenga-Helmantel et  al.

(2014) about the effectiveness of deductive, inductive, implicit, and inci-
dental grammatical instruction in second language classrooms.
 3  Research Methods: Options and Issues 
   67

Using intact classes, a total of 981 Dutch students in lower secondary educa-
tion learning German, English or Spanish participated in this study. The
design of the study consisted of a pre-test, a series of lessons about the
degrees of comparison (the intervention) and a post-test. The tests involved
both meta-linguistic knowledge and production of the grammatical struc-
ture. Differences in students’ test scores according to instructional forms
were examined. The findings indicated that any kind of grammar instruc-
tion, whether deductive, inductive, implicit or incidental, is more effective
than no grammar intervention/exposure.

Questionnaires
Questionnaire studies are very popular, and there are many examples of
this type of study, because they are relatively easy to administer, they can
be used in a time-efficient manner with large numbers of participants,
and they produce data which is relatively easy to analyse.
Constructing an effective questionnaire can be quite complicated and
time-consuming. Table 3.2 identifies some key points to consider when
constructing a questionnaire for a research project.
Especially popular is the Likert-type questionnaire, invented by psychol-
ogist Rensis Likert (1932), where responses are made on a continuum. It
is important to remember, however, that this data is ordinal rather than
numerical, that is, the data produced are not interval, and equal intervals
cannot be assumed. It is not sensible to say, for instance, that someone
who gives a rating of “4” for an item is twice as much in agreement as
someone who gives it a “2”. For this reason, nonparametric tests (e.g.
medians rather than means, Spearman’s rho rather than Pearson’s prod-
uct–moment correlations, and Mann–Whitney U or Kruskal–Wallis tests
of difference rather than t-tests or ANalysis Of  VAriance (ANOVAs)) are
appropriate for this type of data.
Other important considerations include the language the question-
naire is to be presented in—The students’ L1? The target language? The
medium of instruction? These are important considerations, since it is
obviously important that the participants understand what they are rat-
68  Developing Language Teacher Autonomy through Action Research

Table 3.2  Points to consider when constructing questionnaires

Questionnaire construction: points to consider:
1. Keep items short and simple in order to minimize the possibility of
misunderstanding.
2. Use simple, natural language and avoid technical terms or jargon which
not everyone may understand.
3. Avoid nonspecific words such as “good”, “often” which may mean
different things to different people.
4. Avoid words with more than one meaning.
5. Avoid loaded terms such as “democratic”, “free”, “dishonest”, “cheat”,
and so on.
6. Do not ask leading questions, for example, “Surely you agree that…?” or
“Isn’t it only fair to think that…?” since the very wording of the question
may bias a response.
7. Avoid negative constructions, for example, “I don’t use many strategies…
agree/disagree” since such an item can be difficult to respond to
appropriately.
8. Avoid the English construction “Would you mind…?”, since this is also
difficult to answer in such a way that the meaning is clear.
9. For the same reason, avoid tag questions, since the English convention for
answering such questions is often contrary to that of other languages, and
the meaning taken may be quite the opposite of the intended one.
10. Avoid double-barrelled items such as “I learn vocabulary by watching
movies and listening to songs” since respondents may well do one but not
the other, thereby introducing ambiguity.
11. Avoid pluralisms such as “parents”, “children”, “students” since a single
answer may not cover all cases.
12. Avoid culturally biased questions.
13. In order to avoid a response set where participants mark only one side of a
rating scale, it is a good idea to include some items which focus on a
negative aspect, for example, “I believe I am a poor language learner”.
These items will need to be reversed during analysis.

ing. At the same time, many classes are not monolingual, and it may not
be practical to translate a questionnaire into all of the participants’ lan-
guages. This is an issue which requires careful thought and justification
when describing the methodology. Table 3.3 contains an example of a
Likert-type questionnaire.
Likert-type questionnaires typically present the respondent with
a list of statements which they are asked to rate according to their
personal opinions. Rating scales typically range from 1–3 (if fine
 3  Research Methods: Options and Issues 
   69

Table 3.3  Reading strategy questionnaire

The following questionnaire contains some of the strategies which students
report using in order to assist the development of reading skills in the target
language. Please read the following strategy items and grade each one
according to the frequency with which you use it
1. very low  2. low  3. medium  4. high  5. very high
1. I read extensively for information in the target language
2. I read for pleasure in the target language
3. I find reading material at my level
4. I use a library to obtain reading material
5. I first skim read a text then go back and read it more carefully
6. I look for how a text is organized and pay attention to headings and
sub-headings
7. I make summaries of what I read
8. I make predictions about what I will read next
9. I guess the approximate meaning by using clues from the context
10. I use a dictionary to get the exact meaning

distinctions are not required) to 1–7 (if the more fine-tuned distinc-
tions are required). Occasionally scales may ask for an even wider
range, though this is not common. Most scales employ an uneven
number of rating options to allow respondents to opt for a neutral
choice. Others deliberately employ an even number of options in order
to avoid respondents opting for the easy neutral choice. Judgement is
required to decide which of these options is most suitable for a par-
ticular research purpose.
Other types of questionnaire include the checklist questionnaire.
Such a questionnaire is closed-ended because the participant has to
tick only relevant items which could involve ranking, classifying,
ordering, and so on. Table 3.4 provides an example of this type of
questionnaire.
Closed-ended questionnaires are relatively objective, in as far as they
are not as dependent on researcher selection and interpretation as
may be the case with more qualitative methods. For these reasons,
closed-ended questionnaires can be a useful research tool, especially
if used as a base for further exploration, although it is necessary to
bear in mind, acknowledge, and make allowances for their potential
limitations.
70  Developing Language Teacher Autonomy through Action Research

Table 3.4  Which of the following do you normally do when you learn vocabu-
lary? Please tick the top three
1. _____Writing the word several times
2. _____Copying sentences from the dictionary
3. _____Using the word in a sentence
4. _____Using visuals to understand the meaning
5. _____Drawing pictures of the word if applicable
6. _____Finding synonyms or antonyms
7. _____Creating a thematically related word list
8. _____Finding the meaning in a dictionary
9. _____Trying to guess the meaning from the context
10. _____Asking its meaning from others
11. _____Elaborating on its meaning
12. _____Analysing the word structure

Open-Ended Questionnaires
Open-ended questionnaires are ways of collecting written sets of data
by asking the participants relevant questions. Such questionnaire designs
can be used when it is not possible to arrange individual interviews with
the participants. The questions can be delivered through emails or given
to the person on paper. If delivered through emails, there is also a chance
of asking for follow-up complementary data for unclear points. Also, the
researcher will not have to transcribe the responses as would be the case
with handwritten responses. This could help save time and energy for
other issues in the research.
A potential problem with this form of data collection is that if
the participant does not find time to provide complete information,
incomplete data sets might result, which would impact the reliability
of the findings. The results drawn from these sets of data might there-
fore be misleading.
Another potential problem might be the risk of asking leading ques-
tions, which might bias the participants in the direction of giving par-
ticular answers to the questions. For example, questions such as how did
you benefit from this activity? implicitly suggest that the activity was good.
In other words, the question assumes that the interlocutor has benefitted
from the activity, and does not make allowance for any negative feedback.
Instead, neutral or general verbs can be used such as how has this activity
 3  Research Methods: Options and Issues 
   71

influenced you? Such a question leaves the interlocutor enough room for
expressing whatever he/she wants to say.
Another way of avoiding leading participants is not to use confirma-
tory questions such as you learnt a lot from this activity, didn’t you? Such a
question might lead participants to provide a response that would please
the researcher.
In addition to risky question types in open-ended questions, the
researcher also needs to consider his or her relationship with the partici-
pants. It is usually not recommended that a trainer ask questions about
the programme, management, course, professional development, and
so on. This is because social status/power difference might make par-
ticipants feel obliged to provide answers that would please their train-
ers, teachers, boss, or coordinators. To avoid this, anonymous responses
might be required, or, if collected through email, they might be sent to a
colleague who could then pass them on without names. This will remove
the stress the participants might have when answering questions honestly.
The same could be valid for teachers too. When teachers collect data from
their students, they might be unintentionally pushing students not to tell
them their true opinions.
Bearing all of the above in mind, open-ended questionnaires such as
the one in Table 3.5 can be useful. Such a questionnaire provides par-
ticipants with an opportunity to give in-depth answers to the questions,
which in turn offers an opportunity to be able to understand the reasons
for participants’ particular answers in a potentially rich manner. Rather
than selecting a pre-set item in a questionnaire, participants might explain
their actual views on the content of the questions. Such questionnaires
are relatively more creative and in-depth as they provide unpredictable
responses free from the researchers’ biases. The participants are not forced

Table 3.5  Grammar learning preferences

1. What are your current practices for learning grammar?
2. How do your teachers teach grammar in your classroom?
3. What are the grammar teaching practices you like most? Why?
4. What are the grammar teaching practices you dislike? Why?
5. How would you like to learn grammar? Or how would you like your teacher
to teach you grammar?
72  Developing Language Teacher Autonomy through Action Research

to make choices from among the pre-set statements or phrases which

could be biased by the underlying assumptions of the researchers. Such
a set of data could be exciting for a researcher to deal with in that he/she
will read a variety of responses that might touch upon a wide range of the
phenomena in question.
It is also possible to have questionnaires which have both closed-ended
and open-ended questions. These kinds of instruments are very useful
because they ask the participants to make a particular choice first, fol-
lowed by a slot where they can justify their choice. Such a format could
help to elicit specific data on particular issues with underlying reasons. If
there is a particular focus in the research which seeks to identify particu-
lar views or perceptions, participants can be asked to choose between the
two and asked to explain the reasons for their choice. Such a set of data
might make it easy to sort out the data and learn about the participants’
reasons for their choices. This kind of questionnaire might look like what
is provided in Table 3.6.
Potential problems with questionnaires relate to validity (are they
really surveying what they are meant to be surveying? Are respon-
dents sufficiently self-aware and honest?) and reliability (are they
consistent?):

• Can we rely on the fact that respondents are self-aware?

• Have they answered honestly and thoughtfully?
• Have they understood the items?

Table 3.6  Please tick the option you think is most appropriate for yourself and
give a reason
1. Learning vocabulary in context is easy/challenging
because_______________________________________________________________
_______________________________________________
2. Doing extensive reading is beneficial/useless
because_______________________________________________________________
_______________________________________________
3. Learning grammar is a waste of time/important
because_______________________________________________________________
_______________________________________________
4. Learning correct pronunciation is difficult/no problem
because_______________________________________________________________
_______________________________________________
 3  Research Methods: Options and Issues 
   73

• Have they understood the instructions?

• Does respondent A mean the same thing by “often” as respondent B?

Nevertheless, as long as the data are interpreted with care, and, preferably
triangulated with data obtained from a different method (e.g. interviews,
observations), questionnaires are capable of producing useful and inter-
esting data which can provide valuable insights into a research question
and act as a basis for further investigation. These insights may be difficult
to obtain by any other method.
Because of their convenience and their potential to generate interest-
ing data which is relatively easy to analyse, questionnaires are a popular
choice for researchers. Examples of studies which employed questionnaires
include the quantitative section of the writing skills study (Griffiths 2016),
the styles and style stretching study (Griffiths and İnceçay 2015), and the
dynamic strategy development study (Griffiths 2015) described above.

Observation
Observation of students and other teachers or having them observe one-
self while teaching can be a useful way of obtaining data. When observ-
ing, it is useful to use observation rubrics or checklists and to take notes
which can provide a basis for later feedback and help to ensure potentially
important details are not overlooked or forgotten. Observation tools can
change according to the focus of the study. For example, if the purpose of
the study is to investigate questioning techniques:

• When focusing on students, a seating chart for each student can be

useful.
• If the focus is interaction between students and the teacher, arrows can
be drawn to show the direction of the questions in order to identify the
dominant patterns of interaction, which could provide implications
for teachers and students.
• If the focus of the observation is on the teacher with special emphasis
on the type of questions he/she is asking, questions can be written or
recorded (e.g. by video or audio recorder), which would provide a pat-
tern when analysed.
74  Developing Language Teacher Autonomy through Action Research

Observations can also be done on the basis of time intervals. The observer
can write what the teacher does every five minutes. This could give a clear
understanding of the teacher’s time management skills and the variety of
activities within the lesson period.
Another way of getting information from an observation could be
staging the lesson and writing about what the teacher does, for example,
introducing the lesson with a warm-up activity, asking students to form
groups to discuss, providing input on the board, asking students to write
what is on the board, and so on. These could give a clear outline of a les-
son to analyse further the teachers’ preferences. Important questions to
consider:

1. Who will be observed?

–– another teacher (use of board, interaction patterns, classroom
management, giving instructions, use of language, etc.),
–– one student, a group of students, all the students (asking ques-
tions, giving responses, particular behaviours, peer relations, col-
laboration, etc.)
2. Will the observer be
–– studying with students during the instruction (as a participant )
–– observing from a particular place without interaction with stu-
dents (as a non-participant)
3. How will the observation findings be documented?

–– checklists prepared for the specific task

–– rubric for the specific task
–– detailed note-taking
–– summarized notes
–– video/audio recording
Potential problems with using observation as a research method
include the phenomenon of observer paradox (Labov 1972), which
expresses the idea that, although an observer may go into a situation
in order to observe natural behaviour, the very fact of being there is
 3  Research Methods: Options and Issues 
   75

likely to change normal patterns of behaviour. This is also known as

the Hawthorne Effect (Landsberger 1958) after the place where it was
first documented.
Another problem is the idea that much of learner behaviour (perhaps
the most important behaviour) is internal. It is essentially cognitive, and
it occurs inside the head, and is therefore unobservable. Even affective
reactions and social relationships, which may affect cognitive function-
ing, may not be readily observable, especially with older learners who
have often learnt to hide such feelings. Nevertheless, observation can pro-
vide some interesting insights.
Observational data do not help the researcher to understand inner
feelings, attitudes, perceptions, and beliefs. Rather, they are carried out
to learn more about teaching practices, classroom management styles,
and interaction patterns in the classroom.
For example, we might want to investigate interaction patterns used
to locate the kinds of questions asked by teachers and students as well
as initiators of questions. In this case, we need to develop some seating
plans of the classroom to identify who is asking questions and who is
answering. For practical reasons we may draw arrows that show who is
interacting.
Another topic of observation could be teaching practices. For example,
we can easily observe the kinds of exercises and activities offered, the
types of techniques used, and the language of instruction. For this we
can take thick notes that exemplify the focus of observation. Later we can
categorize and count the frequency.
Since frequency alone may not suffice to understand particular
reasons for specific practices, the researcher may need to do follow-
up interviews or post-observation meetings to clarify why particular
activities, techniques, and instructional preferences are used. The data
collected through these two tools could complement each other and
help the teacher gain deeper insights into the phenomenon under
investigation.
Another interesting topic that could be investigated though observa-
tion is language use. Language use of learners and teachers can be recorded
76  Developing Language Teacher Autonomy through Action Research

or written to provide a very valuable source of data. For example, if the

research questions are:

• Which language does the teacher use more than another?

–– Observing and calculating the duration of the use of each
language
• For what functions does he/she use language as he/she does?
–– Taking notes of the functions of the particular language use in
each case.
• How does the teacher move between the languages?

–– Taking notes of whether the teacher code-switches (changes the

language) or code-mixes (integrates words into one language
from another or vice versa)
Let us assume the researcher found from observation that the teacher
used the native language more than the target language. Then, the follow-
up interview question can be:

• Why do you use the native language more than the target language?

If appropriate, the teacher can be told the result of the observation of

his or her lessons.
Again, let us assume that the teacher uses the native and target lan-
guage for particular purposes: the native language when he/she wants to
give instructions and the target language when he/she wants students to
practise it. In the post-observation meeting, he/she can be given informa-
tion about the finding and asked why he/she prefers this.
Similarly, let us assume that the observation showed that the teacher
used code-mixing more than code-switching. We could ask why he/she
mixes words from one language into the other. These follow-up questions
might create depth and breadth in the data collected through observa-
tion. In this way not only scores, numbers, or percentage about the class-
room practices are given, but also these tendencies are supported through
the reported justifications and comments made by the teachers. In this
sense, observational data is supported by interview data so that it yields
the full picture of the issue.
 3  Research Methods: Options and Issues 
   77

The same could be done with learners. After observing particular

learners’ classroom actions and behaviours, follow-up questions can be
asked, by which they can give rationales to account for why they do what
they do. As a teacher, observing other teachers or students could be an
empowering tool for exploration and reflection if the research process
is conducted in a way that addresses the purpose of the study and with
questions that can achieve this purpose. An example of a study which
used observation is the one by Soruç and Griffiths (2015) which explored
the role of identity in the adoption of spoken grammar.

This article reported on a study that investigated issues involved with the
teaching of features of spoken English, sometimes called spoken grammar,
including the use of vague language, placeholders, lexico-grammatical
units and ellipsis. Materials focusing on four spoken features were prepared
and presented over a period of 2 months to 19 students aged 18–20 prepar-
ing to enter a private university in Istanbul, Turkey. During the lessons and
the tests, students were observed and video recorded and field notes were
kept. It was found that although some initial uptake of the target spoken
grammar features was evident at the time of the post-test, little had been
maintained by the time of the delayed post-test three weeks later.

Interviews
Interviews can be a very useful way of providing qualitative informa-
tion. However, conducting interviews is not as easy as it might at first be
assumed. It is important to ask non-leading, unambiguous, and concise
questions. Issues include:

• interviewee selection
–– random (e.g. every third participant)
–– convenience (i.e. they are chosen because they can be conveniently
contacted, whether face-to-face, by email, by phone, etc.).
–– purposive (i.e. they are selected for some particular reason, per-
haps because they fit into a particular age group, because they
answered questions in a way that the interviewer wants to find
out more about, etc.)
78  Developing Language Teacher Autonomy through Action Research

• is it better to interview individuals or focus groups?

• focus should be on specific questions
• the aim is to investigate participants’ insider perspectives on what they
do and especially why they do the things they do

Interviews can be:

1. Structured—the questions are decided beforehand and only those

questions are used.
2. Unstructured—the interview does not proceed according to a pre-­
planned schedule, but according to issues which arise spontaneously.
3. Semi-structured—the interview begins with a set of pre-planned
questions, but other issues or insights are allowed to be included as
they arise.

The following interview schedule was designed for use with learners in a
classroom following a particular lesson (Table 3.7).
Consider:
–– Whether the questions are good? Why? Why not?
–– Which question(s) is/are leading the learner to a particular response?
–– Is there any evidence of researcher bias?
–– Are any of the questions ambiguous?
–– What other questions can you ask?
–– Do(es) the participant(s) feel under pressure to giving a particular
answer?

Table 3.7  Sample interview schedule

Interview schedule
1. Can you talk about what you expected before the lesson?
2. What happened during or after the lesson?
3. What critical issues did you notice during the lesson?
4. How did you feel about the lesson in general?
5. How has it influenced your learning?
6. How did it contribute to your motivation?
7. How do you think the lesson could be improved?
8. How did your classmates feel about the lesson?
 3  Research Methods: Options and Issues 
   79

Problems: In the interests of transparency, interview data should be tran-

scribed, but this can be time-consuming and introduces a subjective ele-
ment in that the transcriber has to interpret what he/she hears. Using a
programme which transcribes speech directly into text may help with this
difficulty.
An example of a study which used the interview technique is the one
by Polat (2013) which investigates what language experience interviews
can tell us about individual differences.

In order to investigate the impact of individual variables on language learn-

ing, rather than looking at isolated variables, this study took a top-down
approach by starting at the level of students’ overall experience of lan-
guage learning, using language experience interviews. During these inter-
views, students were asked to reflect on general and specific aspects of
their language experience. Interviews were conducted with 12 students at
an American university. Qualitative analysis of the interviews indicated that
more successful students seemed more focused on learning grammar and
meaning construction, whereas les successful students were more focused
on memorization and vocabulary.

Introspection
Introspective studies provide an insider perspective on students, other
teachers, coordinators or administrators who report on

• what they do
• why they do it
• what they believe
• what they are or were thinking

A common technique is to get participants to do a particular task (such as

a comprehension or an inferencing exercise) and to vocalize their thought
processes while they are doing it. This is often called “think-­aloud”.
Although introspection can produce some interesting results, care needs to
be taken with interpretation, since the reliability of this method is some-
times questioned because of the high level of subjectivity, and uncertainty
80  Developing Language Teacher Autonomy through Action Research

over the degree to which individuals may be capable of accurately vocaliz-

ing their mental activity. An example of a study which used this technique
is the one by Hu and Nassaji (2014) which investigates the use of lexical
inferencing strategies by more and less successful inferencers.

Using think-aloud procedures with 11 Chinese learners of English, this study

explored learners’ inferential strategies as they tried to infer word mean-
ings from context, the strategies they used to do so, and the relationship
with their success. Twelve types of inferential strategies were found to be
used by all the learners, and two groups of learners were identified: success-
ful and less successful inferencers. The results indicated a number of differ-
ences between successful and less successful inferencers, relating not only
to the degree to which they used certain strategies but also when and how
they used them. Important characteristics of successful inferencers included
frequent use of evaluation and monitoring strategies, a combination of
both textual and background knowledge, self-­awareness, and repeated
efforts to infer the meanings of the target words.

Reports
There are various ways of collecting reports, perhaps the most common
being asking the students to keep a diary, journal, or blog and then col-
lecting and analysing the material they produce. Again, reliability of
results may be threatened by the highly subjective nature of the data. An
example of this kind of study is the one by Lin et al. (2014) about using
blogs to change students’ writing skills and perceptions.

This study attempted to explore the relative effectiveness of traditional

pen-and-paper journals compared with writing blogs during a 16-week
comparative experiment involving two groups of EFL college students. The
experimental group was required to blog daily while the control group was
asked to keep a traditional pen-and-­paper diary. For both groups, writing per-
formance, general attitudes, and perceptions of journal writing were evalu-
ated by means of a writing test and a semi-structured survey. According to the
results, blogging helped the students to achieve a greater improvement in
their writing than did the regular pen-and-paper treatment. In addition, the
blogging group experienced significantly less anxiety in writing than did
the pen-and-paper group, suggesting that the blogging approach is effec-
tive both for developing writing skills and in terms of affective reactions.
 3  Research Methods: Options and Issues 
   81

Case Studies
A distinguishing feature of case studies is that they use individual cases
to generalize about the wider population. However, “individual” does
not necessarily imply just one person: it may refer to a small group, or a
particular class, or even a specific school or larger unit. Case studies aim
to explore the characteristics of individuals, organizations, or groups. The
purpose of such studies is to collect a detailed description and under-
standing of a case. The important thing is that the “case” forms a definable
unit. Case studies are usually longitudinal and often involve qualitative
rather than quantitative data which tends to be “thick” (i.e. there is lots of
it). They often derive theory from data rather than the other way around,
a process known as “grounded theory”.
Strengths of the case study approach include its sense of reality, its
open-endedness, and its flexibility. Weaknesses include the level of sub-
jectivity involved, and the difficulty with establishing reliability and
validity. One of the important aspects of case studies is that they make
use of a variety of data collection methods including interviews, observa-
tions, and documents. This allows for a natural triangulation at the data
collection level.
The four students surveyed in the second section of the study of writ-
ing strategies (Griffiths 2016) described above is an example of a case
study. This study aimed at investigating the way a small group of suc-
cessful students used writing strategies in the hope of obtaining insights
which might be generalizable to a wider population.

Narratives
Narratives are becoming a popular method of researching learning phe-
nomena because they are a way of exploring learner perceptions in a deep
and qualitative way by getting them to tell their own stories. This is,
however, another method which needs to be interpreted with care since
its inevitably subjective nature may threaten reliability.
Teachers could use this type of qualitative research if they want to
understand more about a particular issue through life stories of stu-
82  Developing Language Teacher Autonomy through Action Research

dents. A researcher who wants to study learner motivation could ask

learners to write a narrative of how they started to learn English, and
what happened during these years. The stories will naturally include
several critical incidents which might allow the researcher to get a
deeper understanding of motivational dimensions. The collected nar-
ratives could then be transcribed if necessary and analysed on the
basis of the elements such as time, place, plot and scene, and the
recurring themes in the stories. The initial drafts of written narratives
could then be put into a chronological order by highlighting the set-
ting, characters’ actions, problems, and resolution, if any. As qualita-
tive research studies might have validity problems, it is necessary to
ask the participants to read the narrated stories of their experiences,
which is also known as member-checking. An example of a study which
used narratives is the one by Besser and Chik (2014) which explores
narratives of second language identity amongst young English learn-
ers in Hong Kong.

This article reports on a study which attempted to explore the question of

young learners’ perceptions of themselves as speakers of a foreign lan-
guage. A group of 24 Hong Kong primary-school children (aged from 10 to
12) who are in the process of learning English participated in a study using
a photo-elicitation stimulus. Two distinct types of narrative emerged, which
were labelled “Cosmopolitan” (who seemed to draw on multiple resources)
and “Pragmatist” (who were more uncertain about their future learning
opportunities). Implications of the findings are discussed and suggestions
are made regarding how teachers can facilitate the development of
identity.

Ethnography
Ethnographic studies are those which are situated in a particular cul-
tural context which influences the nature of the study and the particu-
lar findings. A study which might be considered ethnographic because
the context in which it was conducted is a key feature of the study is
the one by Kordia (2015) which explores the pedagogical awareness
of English as a Lingua Franca (ELF) in Greece.
 3  Research Methods: Options and Issues 
   83

This study was carried out in Greece with 12-year-old students (N = 22) in a
6th grade class. Although English is not widely spoken in Greece, the
researcher found that her students often needed to use English while play-
ing internet games or chatting online with foreign friends, and sometimes
they found this difficult. Because of these difficulties, the researcher wanted
to promote the use of English as a lingua franca, and to develop what she
called an ELF-­aware pedagogy. The researcher concludes that it is essential
to consider the learners’ background and to engage them in real ELF
communication.

Phenomenology
Phenomenology is not a new approach, but it has recently become popu-
lar. As the name suggests, phenomenology is the study of a particular
phenomenon (e.g. un-motivation) in a given situation. A study which
might be seen as exploring the phenomenon of learner autonomy (and,
therefore, to be able to be described as phenomenological) is the one by
Güzel (2015).

As Güzel comments, the Turkish education system, where the study is situ-
ated, “is based on traditional teaching methods which ...hinder autono-
mous language learning” (p. 189). In order to investigate the issues involved,
she used both a frequency-scale questionnaire as well as open-ended ques-
tions with 53 preparatory school students and 10 instructors. She found
that, although instructors tried to promote learner autonomy, students
were generally exam-­oriented and lacked autonomy.

Grounded Theory
Introduced by Glaser and Strauss (1967), a grounded approach to theory
generation has become very popular, and involves taking a bottom-up
rather than a top-down approach. Using this approach, the data is pro-
gressively coded by means of an iterative process of constant comparison
according to different phases (Strauss and Corbin 1998):
84  Developing Language Teacher Autonomy through Action Research

• Open—at this stage, the text is examined for salient themes.

• Axial—once themes have been identified, they are grouped around
conceptual axes.
• Selective—at this stage, a core theme which over-arches the others is
selected.

Although not all studies which are described as “grounded” are con-
ducted strictly according to this paradigm, a key feature is that the themes
emerge from the data rather than being pre-determined. This method is
particularly useful for analysing narratives, diaries, blogs, and so on, where,
by their nature, it is impossible to predict the data which is likely to emerge.

Mixed Methods
Since it is often impossible to gain a broad view of any particular phe-
nomenon on the basis of any one method, mixed method studies have
become increasingly popular. By using more than one method (e.g. a
questionnaire followed up by interviews) we can obtain triangulation.
There are many examples of this kind of approach, including the study
of writing strategies by Griffiths (2016) described above, which employed
a questionnaire followed by a case study approach with a small group of
individuals; and the study of the spoken grammar of English by Soruç
and Griffiths (2015) also described above which used observation fol-
lowed by interviews.

Triangulation
Triangulation in action research involves obtaining multiple perspectives
on the problem/question/puzzle/issue being investigated. This can be
done in several ways:

• We can compare data obtained by a variety of methods, using, per-

haps, both quantitative (e.g. a questionnaire or a quasi-experimental
design) and qualitative (e.g. interview, observation or think-aloud)
paradigms to see how similar or different they may be.
 3  Research Methods: Options and Issues 
   85

• We can employ multiple sources for obtaining our data (e.g. students,
parents, teachers, school authorities) to obtain multiple points of view.
• We can use multiple analytical tools, including descriptive (e.g. mean,
median, mode, frequency, percentage) and inferential (e.g. correlation,
difference, effect size), in order to determine whether the results obtained
from one analytical method is similar to or different from another.

Documentation and Artefacts
Depending on the method chosen, it may be important to collect and
keep a variety of documents and artefacts from teaching, such as

• lesson plans
• teaching materials
• student work
• journals
• pictures
• tests and results

If these artefacts are kept systematically, the data that they contain can all
be used as the basis for a research project.

Piloting
As busy teachers, it is easy, in our rush to get things done, to overlook the
importance of piloting (trying out) our research plan and/or instruments.
This is, however, a very important stage, and can actually save time in the
long run, since, although we may spend a lot of time, for instance, con-
structing a questionnaire, it is impossible to know how it will be when
we actually use it. If it is piloted, perhaps with a small group of non-­
participants, any defects will become obvious and can easily be corrected.
If, however, we go ahead and use it with the target students and then find
there is some flaw we had not thought about, we will have wasted a lot
of time and trouble, and it will not be possible to use this data or to use
the same group again.
86  Developing Language Teacher Autonomy through Action Research

Questions to Consider
1. What is the difference between quantitative and qualitative research?
2. What is the difference between cross-sectional and longitudinal
research?
3. What are the essential characteristics of experimental research?
4. How is quasi-experimental research different?
5. What are the benefits and the problems of questionnaires?
6. Explain observer paradox.
7. What are the different kinds of interviews?
8. What are the strengths and weaknesses of case studies?
9. How reliable do you think introspection, narratives, or journal studies
are?
10. What can we learn from ethnographic studies?
11. What can we learn from phenomenology?
12. What are the advantages of grounded theory?
13. What are the advantages of mixed methods?
14. How can you ensure that useful documentation is kept?
15. Why is piloting important?

Example: An Action Research Study on Classroom Culture:

Methodology

In order to obtain more than one perspective on the research questions for
the study on classroom culture introduced in Chap. 2, a mixed method
approach was chosen in order to triangulate one set of data with another
and thereby obtain a more in-depth view. The procedure involved first of
all constructing a questionnaire. As a first step, the issue of politeness was
discussed in class, and students were asked to provide ideas about their per-
ceptions of polite and impolite behaviour. Following discussion, students
wrote these ideas on pieces of paper which were handed in anonymously.
Based on this input, a questionnaire was then constructed using a five-
point Likert scale format, ranging from 5 = very impolite to 1 = not at
all impolite. In addition, in order to obtain qualitative data, a column
was provided where students were asked to comment and explain their
opinions.
In the interests of preserving anonymity if students chose to remain
anonymous, providing their names was optional, and in order to ensure
 3  Research Methods: Options and Issues 
   87

consensual participation, students were asked to sign agreement to using

the data for research or publication purposes at the bottom of the ques-
tionnaire form.
The first version of the questionnaire was initially trialled with a class,
after which adjustments were made before it was given to a different class
(see below). The questionnaires provided by the second class were used
for the purpose of analysis.

Classroom Culture Questionnaire

Gender: M F Nationality:_________________
 lease mark the following behaviour according to how impolite you
P
think it is. Please also add a comment explaining your opinion.
 5 = very impolite  4 = impolite  3 = no strong opinion
2 = not impolite  1 = not at all impolite
Comment (please
Item Behaviour Rating explain your rating)
1 Coming late
2 Using a phone
3 Talking while someone else is talking
4 Not paying attention
5 Sleeping
6 Eating or drinking
7 Using inappropriate terms of address
8 Using bad language
9 Wearing inappropriate clothing
Any other behaviour you consider
impolite

I agree to the use of this data for research or publication purposes:

Tasks
1. Which type of research design will you employ to address your research
question? Justify your choice
2. Construct and pilot an instrument suitable for your study (e.g. a ques-
tionnaire, observation or interview schedule, test)
 4
Thinking About the Context: Setting
(Where?) and Participants (Who?)

In this chapter, we will highlight the role of the contexts in which teachers
are working, since the feasibility of any study depends on the factors cre-
ated by the specific contexts. For instance, if working with teenagers, it is
not useful to consider issues related to young learners (unless there are plans
to collaborate with a colleague who teaches at that level). Similarly, the
books which must be used, the exams for which students must be prepared
are all contextual factors which may constrain the topics which are practi-
cally able to be researched. Likewise the location: in a city school, there may
well be more traffic and less playground space than in a rural school; in a
place well-known for its art, music, scenery, historical buildings or wildlife,
there may well be access to a resource that is simply not available elsewhere.

Research Setting
Research setting is the specific location (classroom, school, institution, etc.)
where research is conducted. The major reason for a detailed description
of the setting of the research is to make the readers understand the condi-
tions under which the data were generated and interpreted. Knowledge

© The Author(s) 2017 89

K. Dikilitaş, C. Griffiths, Developing Language Teacher Autonomy
through Action Research, DOI 10.1007/978-3-319-50739-2_4
90  Developing Language Teacher Autonomy through Action Research

of the setting allows readers to make deeper interpretations about the

meanings of the results too. Setting can generally be described with refer-
ence to four elements:

Place

Places can be seen as geographical location and sociological location. The

former refers to description of where the school, institution, or classroom
is located, whereas the latter refers to the place of the school among oth-
ers such as overall achievement level, socio-economic level, and success
stories. In addition, the type of institution needs to be specified: is it a
pre-school, a primary school, a high school, a vocational college, a uni-
versity? Is it a state or a private institution? All of these factors will affect
the interpretation of the data.

People

Administrative staff, teachers, and learners are key people in a particular

situation. They are critical elements that shape settings in a particular
way. Their educational backgrounds and professionalism influence the
quality of instruction. Some other factors to consider about people in the
location might include:

• how teachers and students are managed by the school administration

• how teachers and students are recruited or accepted to the school
• how teachers and students are treated in the school organization
• how much teachers are paid
• the extent to which teachers and students are supported for
development
• size of classes

Programme

The particular programme students are working on is also an important

piece of information for readers trying to understand the study. Results
 4  Thinking About the Context... 
   91

may need to be interpreted quite differently according to whether the

participants are young learners, high school pupils, university prep school
students, or postgraduates.

Things

This category might include the physical conditions of school build-

ings and classrooms, furniture (desks, seating, etc.), labs, technological
facilities, Internet access opportunities, availability of teaching/learning
materials (textbooks, etc.), and access to libraries and online materials. A
setting equipped with such facilities could enhance motivation and learn-
ing opportunities and one deprived of these could impact the quality of
learning and teaching.
Places, people, programmes, and things are interrelated elements
that need to be considered as a whole which makes up an organiza-
tional system. These are also specific factors that give readers a general
view of the research setting. In fact, specific details about the setting
are the elements that shape the findings of research or are the underly-
ing causes of why results (high or low scores, particular themes about
experiences, beliefs and opinions) are the way they are. The constraints
or conditions may also help researchers interpret some findings. For
example, two of the mentors of action research in Dikilitaş and Wyatt
(2017) found one of the recurrent themes to be providing psychologi-
cal support for action researchers. They agreed that situational factors
such as teaching overload, no allowance of specific time for engage-
ment in research, lack of background knowledge, and lack of incentives
and funding for doing research could be factors contributing to the
need for such support. In the same study another mentor experienced
difficulty in creating sustainability and motivation. When the situa-
tion was examined, the teacher-researcher project was found to be a
top-down decision which made it compulsory for all the instructors.
In these cases, the situational constraints shaped mentoring strategies
and practices in different ways. The situational factors might therefore
serve as explanations for particular behaviours, preferences, strategies,
and practices.
92  Developing Language Teacher Autonomy through Action Research

Another reason for writing a detailed description of the research setting is

that it creates credibility, so that the reader knows that the research reported
in a paper has been conducted in a real situation which they can compare
to their own when evaluating the results. They can then decide whether
similar results can be obtained in their own settings if they know as much
as possible about the actual setting where the research was carried out.
Reflection and evaluation of findings are also strengthened by detailed
reporting of the setting. If a research study is carried out in a language
programme, what could be reported in the research setting section might
include:

• Type of school (primary, secondary, high school or tertiary)

• Type of programme (modular system, etc.)
• Teachers’ profile (range of experience, ages, nationalities, etc.)
• Professional development opportunities for teachers
• Support for teachers and learners
• Learning and teaching hours
• Workload and responsibilities of teachers
• Types of materials used (coursebooks, online materials, etc.)
• Facilities (technology and extracurricular)
• Testing and assessment practices
• Location of school

It might be necessary to select the relevant items to report in the

research. If, for example, the research is focused on learners, it may not be
necessary to provide information about the status and profiles of teachers
unless specifically relevant. This is for the researcher to decide, but, in
general, we can say that the more information that can be provided, the
more credible the study will appear to be, and the easier readers will find
it to decide if it is relevant to their own situations.

Factors to Be Considered
Situation in action research must be considered as an integral factor
when developing a research plan. Any researchers need to know and
 4  Thinking About the Context... 
   93

explicitly discuss the specifics of the setting in which the research will
be carried out. This is critical for several reasons. For example, any
research has a setting that is interrelated to the research focus and
problem. Setting is where the research focus emerges as an issue to be
investigated in detail by means of the research questions. Therefore,
any findings will be directly related to the characteristics or the con-
straints of the setting. Action research is a situation-specific type of
research which focuses on issues problematized by teachers in their
work places, for instance, a classroom, a school, or an educational
institution. Therefore, the issues will be a natural part of these situ-
ations. When issues teachers investigate in their own settings are
considered, they usually focus on a specific classroom practice, their
teaching, classroom management, or curriculum and syllabus issues.
These investigation topics reveal the role played by the setting because
issues are chosen for investigation as they appear in a specific setting.
In different settings, different research points could emerge. For exam-
ple, imagine two different teachers are problematizing motivation in
their classrooms and they arrive at different results. The major reason
for this could be the classroom situation they work in. While in one
classroom, there were poor conditions and a low socio-economic envi-
ronment, in the other the conditions and resources are abundant. It is
not surprising that the factors that affect motivation in different class-
rooms are different. Situations show the uniqueness of each classroom.
What makes action research so powerful is that no two teachers might
be able to reach the same conclusions on similar issues of research.
There is diversity in each classroom, which shows the need for and the
role of action research engagement by teachers.

Types of Action Research Settings

It is also important to consider the following aspects when conducting
action research. These could help teachers locate their research for an
international audience.
94  Developing Language Teacher Autonomy through Action Research

National

Teachers work in a country which has particular foreign language teach-

ing and learning policies. Research conducted by teachers will be influ-
enced by the linguistic environment, and this could be key to the topic
selection. The national environment can be seen as having two layers as
follows:

EFL (e.g. Turkey, Russia, Japan etc.) environments are where language is
taught as a foreign language, usually confined to classroom instruction
with limited opportunities for learning and practice beyond the
classroom.
ESL (e.g. Canada, England, Australia, New Zealand, America etc.) envi-
ronments are where learners have abundant access to opportunities to
practice what has been learnt beyond the classroom. Learners develop
language through active use in the social environment.

Local

Action research is conducted in a local environment where the schools

are established. Therefore, it is necessary to discuss the characteristics
of the local situation with special emphasis on the economic and social
state. The major characteristics may also include the geographical (urban,
suburban, rural) features, the socio-economic status (poor, developing,
developed, and wealthy), and the facilities available to teachers and stu-
dents may vary considerably according to these local factors.

School and Institutional

Schools are settings where action research is carried out with reference
to the current pedagogical decisions (curriculum), practices (syllabuses),
and materials. Action research topics can be generated on the basis of the
emerging problems or questions in this specific situation. It is also impor-
tant to identify and discuss the type of school where research is con-
 4  Thinking About the Context... 
   95

ducted. These could be either a state or a private school, or a university,

high school, secondary school, primary school, pre-school, and so on.

Classroom

It is equally necessary to pinpoint the type of class, such as whether it

is General English, English for Specific Purposes (ESP), English for
Academic Purposes (EAP), content and language integrated learn-
ing (CLIL), English as a medium of instruction (EMI), or one of the
numerous other special purpose types of courses available. Each of these
may well present different puzzles, which, in turn, will require different
research questions when initiating action research. Similarly, discussing
the proficiency levels such as low, intermediate, or advanced could also be
an important piece of situation-specific information.

Participants
Two terms which are often confused when talking about participants are
population and sample, but it is important that their meaning is kept
distinct. The sample is the specific group of participants in a particular
study (e.g. a class of university students in the action research study on
classroom culture we are following in the current volume). Population is
the broader term: it refers to the wider group to which the findings of a
particular study might be generalizable (for instance, in the “classroom
culture” study, Turkish university students more generally, or even stu-
dents internationally).
Participants are key informants of any kind of research. Participants
of an action research could be learners, teachers, administrative staff,
and parents though it is possible to find other people to contribute to
such research. The key role of participants in research is that they are
the source of information on which interpretations are built. In action
research, students in a classroom are the best participants for a specific
research issue problematized by the teacher.
96  Developing Language Teacher Autonomy through Action Research

However, particular students might be selected in the classroom who

can provide responses about a particular issue. For example, interviewing
students who repeatedly failed in the exam will give researchers useful
data if they are investigating the causes and effects of exam failure. In this
case, what successful students might report would not be applicable. In
this way, researchers can add validity to the findings by collecting data
from appropriate groups of participants.
In action research carried out in classrooms, there is need to establish
some rapport with students. Informing the students about the research
and talking about the purpose of research could strengthen the reliabil-
ity of the data as students will be willing to contribute to the research.
Another way of strengthening the reliability is that students are invited
to be co-researchers with the researcher, namely, the teacher. Students
becoming part of the study as researcher rather than merely as data pro-
vider will make them take the responsibility for the research and become
curious about the results. They will be aware that the results will also con-
cern them and they will benefit from the research process either as intel-
ligent beings or by exploring issues that might be interesting for them
too. In many cases, students provide data for researchers, but they never
seem to benefit from such an activity, and this can create resentment and
an unwillingness to participate.
Action research in a classroom could also develop effective rapport
between learners and teachers by bringing them together in trying to
explore and understand issues that they need to learn more about. This
also helps them set up a democratic environment in which both can con-
tribute to the learning and teaching process to varying degrees.
It is evident, then, that students can play a different role in action
research by becoming co-researchers. Rather than introducing them
as participants of research, research teachers can consider them as
­co-­researchers who are studying the classroom issue together with them
as insiders. Since the issue relates directly to them, it is also these students
who can best understand the phenomenon and come up with insights for
deeper understanding. Rather than externalizing students as participants
who provide data, they need to be included in the research study as actual
people who are part of it, although the degree of their participatory roles
might differ, depending on the purpose of research.
 4  Thinking About the Context... 
   97

Allowing learners a role as co-researchers (Allwright 2003) turns the

classroom into a research situation where systematic engagement in
research is carried out at particular intervals. There could be two ways
of engaging in such a practice: integrationist or separatist. In the former,
action researchers can integrate research-related activities into classroom
pedagogy and prevent loss of pace of course flow, while in the latter, they
identify particular times to engage in collaborative research. Through the
integrationist approach, teachers can create meaningful activities which
also lead to language practice and provide learners with learning and
development opportunities. This is an approach also found in the research
process of exploratory practice, termed PEPAs (potentially exploitable
pedagogic activities) by Allwright and Hanks (2009). The integrationist
perspective can also make learners as co-researchers more confident in
participation and contribution, thereby making the research process one
that is not stressful for both. On the other hand, the separatist approach
could make the research something distant from the learning and teach-
ing process in which students may not want to participate. If the ultimate
goal of doing action research in the classroom is to help learners develop
language skills and innovative perspectives for themselves, then it will
not be useful to separate the research process from the syllabus-oriented
course process.

Understanding the Context
Understanding the context of research can be done in different ways.
These include observing classrooms, students, and teachers; initiating dia-
logues with students and teachers; examining documents (lesson plans,
materials, books, syllabuses, complaints, etc.); eliciting written feedback
from students and teachers; and interviewing administrative staff.

Observing Classrooms, Students, and Teachers

By far the simplest and most direct way to gain insights into a context
is to go and observe it. This may involve going into a classroom or series
98  Developing Language Teacher Autonomy through Action Research

of classrooms and watching/listening to what is going on, trying to pick

up the tone of the exchanges, listening to what is said/not said, and so
on. Simple equipment such as an audio recorder or video on a mobile
phone can be effective for record-keeping and is relatively non-intrusive.
Observation has its limitations, however, in that much of what goes on
in a classroom is inside students’ heads, and, therefore, by its nature,
unobservable. Also, the well-known Hawthorne Effect, also known as
observer’s paradox, must be borne in mind, since an observer can change
normal patterns of behaviour merely by being there, meaning that his/
her very presence means the behaviour he/she is observing is not normal
(see Chap. 3).

Initiating Dialogues with Students and Teachers

Context is where students and teachers interact inside and outside class-
rooms. Administrative people also have pivotal roles since their decisions
determine the way teachers and students behave and interact in order to
enable learning. An action researcher, therefore, needs to understand the
experiences of these parties in order to reveal particular characteristics of
the context. The dialogues which researchers might initiate with students
and other teachers in the same school will provide a number of oppor-
tunities to gain deeper insights into the specifics that might relate to the
research question.

Examining Documents

The curriculum documents in an institution could constitute valuable

information about the micro-level decision-making process, which teach-
ers can synthesize with the emerging issues to be investigated. These writ-
ten documents could reveal the contradictory or confirmatory aspects
between the decisions and practices carried out in line with them. Reading
about any written documents relating to student complaints and so on
might provide interesting material.
 4  Thinking About the Context... 
   99

Eliciting Written Feedback from Students and Teachers

It may be that some are more comfortable writing their ideas down rather
than talking face-to-face, which may be threatening. If so, surveys can
provide useful information, or they can write by email or use one of the
many messaging systems available these days which are convenient and
comfortable for all concerned.

Interviewing Administrative Staff

Researchers could also interview decision-makers or designers of the cur-

riculum to understand the underlying perspectives behind the syllabuses
already in use. The interview responses will help action researchers develop
insightful perspectives over the issues under research. These could lead to
deeper understanding of issues problematized by the researcher and sug-
gest ways of dealing with them.
The factors discussed above relating to context of research can be
brought together under four themes: parties involved, academic factors,
learner variables, and location, each bearing an impact on action research
by teachers as set out in Table 4.1:
An action research considering the factors in Table 4.1 is highly likely
to create a comprehensive research focus. Seeing the bigger context will
help the action researcher develop invaluable knowledge that could help
explain and understand the issue in depth.

The Role of Action Researchers

Although much research is carried out by experts, researchers, and
academics in general, action research is usually conducted by teachers
actively teaching in classrooms. From this perspective, action research-
ers are insiders who can develop relevant insights into the issues, while
external researchers may not go into the same depth of understanding
of issues. Although academic researchers could be fully equipped with
research skills and competence and they can conduct robust research
100  Developing Language Teacher Autonomy through Action Research

Table 4.1  Contextual factors to be considered when planning a study

Parties involved Academic factors Learner variables Location
Critical friends Curriculum Beliefs ESL/ESL
Learners/ Syllabus Expectation environment
teachers Materials Needs preferences National
Supervisor(s) Testing and Styles/personality Local
Parents/family assessment Strategies Institutional
Colleagues requirements Age State/private
Administrators Gender Urban/rural
Employers Autonomy Socio-economic
Motivation Primary/secondary/
Affective variables tertiary
Investment
Aptitude
Nationality/culture/
ethnicity
Identity

studies, teachers may lack these qualifications. However, teachers have

longer accessibility to students’ potential feedback and responses with a
continuous mode of observing and understanding issues from different
angles. Teachers’ emerging perspectives could be relatively more mean-
ingful compared to those of academics who might have brief access to
data sources, in the sense that longer engagement and rapport with stu-
dents could provide depth and breadth of issues under investigation.
Action researchers need to look for ways of connecting research pro-
cesses and classroom practices to allow students to spend meaningful
time on engaging in research. What is meant by “meaningful time” is
that learners acquire new skills or language knowledge incidentally or
purposefully. Research in the classroom may well allow for such empow-
ering learning in which learners discover and internalize new language
knowledge. In this sense, teachers and students co-develop new skills and
knowledge, which might make the learning environment more motivat-
ing and exciting as well as create increased learning opportunities. Both
teachers and learners develop autonomy by taking the control of their
own learning through inquiry-based activities involved in the research.
The interdependence created by the collaborative research opportunities
enriches learning in a social context. Action research in the classroom
may help teachers develop new roles as active agents who study with
 4  Thinking About the Context... 
   101

students for the development of teaching practices. These emerging roles

include the following:

Rapport Builder

Teachers as initiators of research in the classroom develop an approach

to co-researching with learners for building rapport with students as co-­
workers on classroom issues. This collegiality established is critical not
only for the research being undertaken but also for future classroom
interactions and collaborations.

Research Designer

Teachers begin to think about ways of developing connections between

the research and the learners. They need to create ways of designing class-
room research and plan the research procedures. This looks like planning
lessons every day, but designing research requires a deeper look into the
process of identifying foci of research, generating data, analysing data,
and interpretation in collaboration with students in the classroom.

Research Leader/Peer

Students could be seen as relatively inexperienced researchers compared

to teachers, so teachers need to take on a research leader role at the ini-
tial stages, especially while students are developing research skills and
perspectives. This new role gives them the responsibility for proper guid-
ing of students in participation and collaboration in research. Teachers
swing between research leader and peer roles as learners develop a sense
of achievement and understanding.

Strategic Collaborator

Teachers seem to be leading the process of researching and then are likely
to be collaborators, but they need to be strategic in leaving the control
102  Developing Language Teacher Autonomy through Action Research

of the research to learners. They might need to collaborate strategically

by scaffolding them at times of conflict and challenge. Similarly, learners
may also take on new roles in the classroom, which might make them
more autonomous and active in learning.

Decision-Maker

By taking part in the research in the classroom, learners are granted the
authority to make decisions about themselves, their materials, as well
as instructional practices. Rather than only being passive recipients of
knowledge provided by teachers, they develop decision-making skills
about what they want, how things can be done by teachers, and how
learning and teaching might inform each other.

Research Collaborator

Since learners become co-researchers with teachers and as they perform

joint-action research, they develop collaborative skills. This new role
makes them more active in several ways: thinking and problem solving,
taking responsibility for action, expressing their own ideas and perspec-
tives, and attempting to promote learning and teaching.

Explorer

Exploring is a process of understanding issues by observing, synthe-

sizing, and assessing with little support from others or with others.
Becoming a learner as an explorer might bring several benefits for the
learners. In this sense, becoming an integral part of the research process
in the classroom urges learners to start exploring things and people in
the context. By the help of research, they develop new skills in learning
by exploring.
 4  Thinking About the Context... 
   103

Example: Setting and Problem

Tackling speaking challenges faced by low-level learners of English through
consultation with students
Elif Başak Günbay and Gülizar Aydemir
When we first had speaking classes in an A1 classroom, we had not expected
that we, the teachers, would be the only ones talking. The first impression
that we had in the first five minutes of the lesson was that we would have
a challenging quarter. However, we couldn’t ignore the fact that speaking
is crucial because it improves students’ communicative skills and only then
can students express themselves. This study focuses on the difficulties stu-
dents (beginners) face with speaking in English.
We tried to explore these challenges from the students’ point of view,
exploring why many second language learners, especially those who don’t
have a strong background in English (like our students in our preparatory
school) feel that speaking in a foreign language is harder than writing, lis-
tening or reading. We also wanted to find out when and at what point the
students should be expected to speak in the target language. To do this, we
prepared questionnaires to find out what impedes their speaking skills.
Analysing the feedback from the questionnaire, we gave a second ques-
tionnaire to ask for solutions from the students. According to the reactions
and feedback, there are two main factors: linguistic and non-linguistic. This
study provides a student perspective on their own challenges and offers
solutions to major problems in speaking. It also answers a question for us:
what should be expected of low level learners in terms of their speaking at
the beginning of their learning experience?

Example: Participants
Team-teaching for teacher training
Nicholas Velde
The two pre-service teacher participants involved in the research included
the author of this article and one other teacher. Both participants were
enrolled in a Master’s degree programme for teaching English as a second
language in an American university. Both teachers were in their mid-­
twenties, one male and one female. The data collected concerned actions
taken in an ESL course offered through a partnership between the univer-
sity and a local non-profit organization offering English tutoring to resi-
dents in need. Twelve to fifteen students attended each lesson, and the
104  Developing Language Teacher Autonomy through Action Research

course was designed to assist the students in speaking with the teachers of
their children about school-related issues, but also included English helpful
for communication in general. (Taken from Velde (2015)).

Questions to Consider
1. What is the difference between setting and participants?
2. Why is context so important in action research?
3. What are the four levels of setting that must be considered?
4. What are some of the ways of going about understanding the context?
5. Which of the four contextual factors noted in Table 4.1 do you consider
most important? Why?
6. What are some of the advantages of viewing students as
co-researchers?
7. Are there any other details of the setting of Günbay and Aydemir’s study
you would like to know more about?
8. Are there any other details of the participants of Velde’s study you
would like to know more about?

Example: An Action Research Study on Classroom Culture: Context

Participants:
There were 35 students who participated in this study. They were in the
fourth year of a four year degree in English Language Teaching (ELT),
so they had already been at the university for some time, and might,
therefore, have been expected to have clear ideas about the standard of
behaviour expected. There were 13 male students and 22 female. Turkish
students were a majority (N = 24), and there were 11 international stu-
dents (from Nigeria, Turkmenistan, Tajikistan, Iraq, Kyrgyzstan, Greece
and Kazakhstan).
Setting:
The students were attending a Sociolinguistics course in the ELT
Department in the Education Faculty at a private university in Istanbul,
Turkey. Since the study was carried out in an actual classroom, where
 4  Thinking About the Context... 
   105

students were mainly concerned with passing their course (especially

since many hoped this would be their last semester before graduation),
the study had a multiple purpose: in addition to the research purpose,
the topic was used to stimulate discussion, and also to act as a model for
research projects in which students were engaged as part of their course.

Task
Specify the context of your study
Write a context section about the environment of your research. Include
relevant information specified below, and any other details you think are
important to your study.
The setting—include as much detail as possible, for example, where
exactly is it located (district, city, country), what type of institution is it (e.g.
state, private, primary, secondary, tertiary), is it co-educational, the socio-
economic level if applicable, and so on.
The participants—how old they are, gender mix, where they are from,
their proficiency levels, what/why they are studying, and so on.
 5
Collecting the Data

In order to conduct quality research, and to be able to discover mean-

ingful answers to our problem/puzzle/question/issue, it is necessary to
collect appropriate and accurate evidence on which to base our conclu-
sions. But before we deal with the technicalities of data collection, we
need to deal with an often neglected but vitally important issue: conduct-
ing research ethically. In fact, this is an essential consideration, in order
to protect all involved: the students, the teachers, the institution, and
even the researcher himself/herself. Although it is very easy to rush into
a study without adequately considering the ethical requirements, a bit of
time spent checking that all is in order from an ethical point of view is
time well spent. This is especially true if publication is envisaged, as good
journals are getting tougher on these requirements. Having said this, it is
also true that ethical “rules” vary from place to place, but, a few general
guidelines may help.

© The Author(s) 2017 107

K. Dikilitaş, C. Griffiths, Developing Language Teacher Autonomy
through Action Research, DOI 10.1007/978-3-319-50739-2_5
108  Developing Language Teacher Autonomy through Action Research

Research Ethics
Getting Permission

It is essential to get permission from two important sources: the institu-

tion and the participants.
Most institutions will have requirements for gaining permission for
research. This may include getting permission from a research commit-
tee, from parents, or from other stakeholders. Failure to comply with
these requirements can have unfortunate consequences.
Permission is also required from participants. Exactly how this should
be done may be set out by the institution, but it should include getting
some form of signed consent from participants.

Informed Consent

It is essential that participants are fully informed about the nature of

the research to which they are contributing. In order to convey this
information:

Will there be a written sheet explaining the nature of the research?

Will it be explained verbally?
Will there be both verbal and written explanations?
Will the explanation/s be in the target language or the L1?

The institution may have requirements for this, in which case, of

course, this is the way it should be done.

Voluntary Participation

Participants have a right to be assured that their participation is entirely

voluntary, and they have a right to withdraw at any time if they do not
wish to be included. Voluntary participation is also methodologically
required to collect reliable data. When people want to participate, they
 5  Collecting the Data 
   109

are more likely to provide honest and accurate responses, which will
increase the reliability of the data.
There are various ways that people might be convinced to participate
in research, for instance:

• Assuring them that they will be informed about the results

• Making sure that they will also learn while providing data
• Telling them that the data they will provide is relevant to what they are
doing

As discussed in Chap. 4, if researchers can make the participants part

of their research, as is the case with action research, this could also lead to
greater willingness to participate voluntarily.

Anonymity

Some research projects are conducted anonymously in order to protect

information which may be considered sensitive or private. Participants
have a right to remain anonymous if that is what they choose, but this
will, of course, limit the degree to which any follow-up studies (e.g. inter-
views) are possible, or the extent to which responses can be matched with
other data (e.g. course grades).

Confidentiality

Where participants do identify themselves, it is essential that all the data

collected are kept confidential and they should be assured about this.
Although it may be tempting to share a particularly interesting piece of
information gathered during the data collection process with colleagues
or others, it is essential to remember the participant’s right to keep per-
sonal information private and confidential. If nothing else, failure to
respect this responsibility may prejudice a researcher’s ability to obtain
further information if the participant in question finds out that his/her
confidence has been betrayed. At worst, such a participant may decide
to take legal action if he/she feels sufficiently strongly about the matter.
110  Developing Language Teacher Autonomy through Action Research

Storage of Data

All the data should be stored in a safe place to prevent any access by
third parties. Many institutions also have requirements about how long
data should be kept, where, and so on, and this requirement should be
observed.

Absence of Threat

All participants should be assured that the information they provide dur-
ing the study will have no effect on them personally (e.g. it will in no way
have any impact on their grades). Another consideration could be to let
them know that the research is a way of learning what they think about
particular issues and finding different ideas and suggestions in order to
improve their learning experience. This could help minimize desirability
bias as students might otherwise tend to provide what they think are
appropriate rather than honest answers.

Professional Distance

An important but often under-acknowledged aspect of ethical behav-

iour is the maintenance of professional distance. In other words, it is
important to maintain and respect the teacher–student relationship at
all times. Crossing the professional lines can be especially problem-
atic where there may be one-on-one encounters, for instance, during
interviews, and this has even more potential for difficulties when power
distance is considered.

Data Collection Options

Once ethical issues have been sorted, there are several different options
for collecting data for action research:
 5  Collecting the Data 
   111

In-class  This might include:

Collecting lesson material (e.g. their writing)

Video or audio recording of student activity
Administering a questionnaire during class time

Data produced in class may be time-consuming, but has the advantage

that the teacher can be sure it is the student’s own, and that all of those
who are present at the time will contribute something.

Out-of-Class  This might include:

Collecting homework (e.g. an exercise designed to check how much a

given lesson topic has been internalized)
Collecting assignments
Collecting other material such as pictures and so on

Data collected out-of-class has the advantage that it does not use valu-
able class time, but the disadvantage is that the researcher cannot be
sure whose work it actually is (e.g. parents often help their children with
homework which may present a distorted picture of the child’s actual
ability). Also, there may well be students who do not hand in homework,
thereby reducing the quantity of data.

Online  Most students these days are very technologically competent and
active, so they may well find filling out a questionnaire or completing
an assignment online more enjoyable than completing the same task in
more traditional ways. Also, from a researcher’s point of view, there are
many online platforms (e.g. SurveyMonkey or Google Forms) which
provide a relatively easy way to construct, distribute, and collect research
instruments.

There may also, however, be some who do not respond well to such
techniques, who do not have such technology available, or who face
112  Developing Language Teacher Autonomy through Action Research

parental restrictions regarding the amount of time they are allowed to

spend online. They might need either to be allowed to opt out or to be
provided with alternative ways (e.g. paper and pencil) to contribute.

Personal Invitation  Especially with more qualitative methods (such as

interviews or think-aloud protocols) it may well be most effective to per-
sonally invite the participants. There are several options for this:

• A one-to-one meeting: this can provide quality, individual data, but

can be time-consuming and difficult to arrange.
• A focus group: this can be a way of gathering more data more quickly
than a one-to-one approach, but lessens the individual perspective,
and tends to mean more exposure for the more extroverted members
of the group while the ideas of the quieter ones are
under-represented.
• Face-to-face: this might, perhaps, involve greeting passers-by (e.g. on
the street, in a shopping-centre, in a restaurant, on a university cam-
pus) and asking questions. Although this method may have the advan-
tage that it is more truly random than other data collection options,
where participants are pre-selected, a researcher may have to cope with
a high rejection rate from people who do not wish to participate.
• By phone: this may involve “cold calling” randomly selected numbers
(which often has a high rejection rate), or pre-arranging calling times,
which can be time-consuming and problematic if it involves calling
across time zones. Depending on the phone system being used, it may
also be expensive.
• By Skype: Skype is relatively cheap (if calling to a landline), or even
free (if computer to computer). Also, there is a choice of just a voice
call or a video call. Arranging suitable times can, however, be
­problematic (especially if widely varying time zones are involved) and
the quality of the calls can be variable.

There are, of course, advantages and disadvantages to all of these

options, so the best one will depend on context. But however it may be
done, it is important that all details (such as where, when, with whom)
   
5  Collecting the Data  113

are clear and that all ethical considerations (e.g. that attendance is volun-
tary, that there is no threat) are clarified.

 ollecting Data from Multiple Sources:

C
Triangulation
Collecting data from multiple sources is known as triangulation, which
is an important aspect of the data collection process. Triangulation in
the data collection stage can be achieved by integrating data from dif-
ferent sources such as surveys, documents, interviews (written or oral),
and observation. For example, an action research study might have three
questions as follows:

1. Can I encourage my student to create authentic writing outside the

classroom?
2. What kinds of things do they write outside the classroom?
3. Do my students benefit from authentic writing outside the

classroom?

An action researcher could refer to several sources of data to answer

these questions. Table 5.1 displays potential data sources for each ques-
tion above, which could add to the strength of the findings that are cor-
roborated by different sources of data.
More can be added to this list, but three different sources of data
for each question would probably be sufficient. Sometimes, the same
source of data for different questions can be used. By seeking to answer
each of the research questions from multiple sources, the level of cred-
ibility of findings will be enhanced, which makes the readers more con-
vinced of the results.

Table 5.1  Data sources to address research questions

Question Source 1 Source 2 Source 3
1 Personal reflections Questionnaire Writing tasks
2 Interviews Writing tasks Feedback from learners
3 Grades Interviews Field notes of teacher
114  Developing Language Teacher Autonomy through Action Research

Some Useful Data Collection Materials

Questionnaires

Those who are involved with helping researchers to set up a study will
know that often the most difficult stage is finding or creating useful col-
lection materials. This is often (though not always) a suitable question-
naire, which will serve to gather the basic data. Pre-existing questionnaires
may have disadvantages (e.g. they may not always be contextually appro-
priate for the proposed research environment), and these should always
be carefully researched and acknowledged. But they have also often been
extensively examined statistically and widely used in other studies, which
can serve as comparisons. Since constructing and validating a new instru-
ment can often be an intimidating prospect for inexperienced research-
ers, existing questionnaires might be worth considering, as long as they
are chosen and justified with care, and, where necessary, adapted to the
new research context. Some well-known questionnaires which action
researchers might like to consider are:

• FLCAS (Foreign Language Classroom Anxiety Scale), first published

in 1986 by Horwitz, Horwitz and Cope. The FLCAS has been used in
many studies since, which may be interesting to compare with new
studies. Various updates are also available.
• BALLI (Beliefs about Language Learning Inventory). This was first
published by Elaine Horwitz in 1987, and it has also been used in
many studies since. If access to the original version is not available, it
may be possible to find it as an appendix to some of the other
studies.
• LSPQ (Learning Style Preferences Questionnaire). First published by
Joy Reid in 1987, this is another survey which has been used in many
studies, and which might be useful for investigating student styles,
perhaps in connection with another variable, such as gender or
nationality.
• SILL (Strategy Inventory for Language Learning). This inventory was
first published by Rebecca Oxford in 1990. Again, this inventory has
 5  Collecting the Data 
   115

been used in many studies. If access to it in the original book is not

available, it may be found in other places, such as the adapted version
in Griffiths’ (2003) thesis.
• SRVoc (Self-regulation in Vocabulary Acquisition). A more recent
addition to the language learning questionnaire repertoire is the SRVoc
(2006). This is a survey which might appeal to those who are inter-
ested in the self-regulation movement and also in vocabulary learning.
It might also provide a useful model for those who are interested in
extending their research into other language development areas such as
grammar.

In addition to these well-known questionnaires, the authors of this vol-

ume would like to provide several other survey instruments that research-
ers might like to consider, either in their current form or adapted to suit
their own interests, contexts, and requirements:

The ELLSI (English Language Learning Strategy Inventory)

The ELLSI was first used in Griffiths’s (2003) PhD thesis. This updated
version has been produced to reflect the increased use of technology
as learning strategies in the years since. In addition, the original ELLSI
was written to be used in an ESL (English as a second language) envi-
ronment, and some items (for instance, about noticing the language
used in the environment) were not relevant for EFL (English as a
foreign language) contexts. This updated version is intended for a
wider and more “tech-savvy” context. Nevertheless, those who are
interested in considering the ELLSI for their own studies should still
consider its suitability for their own context. It would be a very good
idea to try piloting it with a group (preferably of 30 or more) that are
not intended for inclusion in the main study. Reliability could then
be checked (see Chap. 6 for how to do this) and perhaps any unreli-
able items considered for elimination. Any items which students
might add to the “other strategies” box might also be considered for
inclusion, which would help to strengthen the content validity of the
study. The ELLSI is easily adapted to use with teachers if their percep-
tions are of interest, perhaps to compare with students’ perceptions.
116  Developing Language Teacher Autonomy through Action Research

The Language Skills Development Survey

The Language Skills Development (LSD) survey was first developed by
Griffiths (2004) in order to explore the strategies used by Research
Methods students to develop the skills they needed to complete their
course. The survey was designed to give equal weight to each of the
skills (ten items each). Since the development of skills is an important
aspect of many language courses, this could be a useful survey for
action researchers to employ in their own teaching contexts, perhaps
with some adaption to suit specific situations.
The Writing Skills Strategies Survey
If more of a focus on a particular skill is required, a survey such as the
Writing Skills Strategies (WSS) might be useful. This consists of 20
writing strategies, providing a more thorough coverage of the target
skill than the LSD (with only ten items) which students are asked to
rate from 5 = very high to 1 = very low. As always, it is for the researcher
to judge whether to ask students to provide names and/or ID num-
bers, according to the requirements of the situation and the research
goal. Also, any other individual details (e.g. age) are a matter for case-­
by-­case judgement. The WSS could be used as a model for researchers
who wanted to study other skills (reading, listening, and speaking).
The Inventory of Language Learning Styles
The Inventory of Language Learning Styles (ILLS) was a survey used by
Griffiths and İnceçay (2015) to explore the styles used by their stu-
dents and the relationship of their style preferences to successful course
outcomes. Since none of the existing style instruments seemed to give
a sufficiently broad coverage of possible style types, a wide range of
these types was gathered from the literature and included in the survey.
If a more in-depth look at any of these types was suggested by the
results of the survey, it would be possible to follow up such insights,
perhaps with interviews.
The English Language Motivation Questionnaire
The English Language Motivation Questionnaire (ELMQ) was used by
Griffiths and Özgür (2013) to explore students’ motivation and
 5  Collecting the Data 
   117

correlate this with their success rates, using the traditional dichotomies
(integrative/instrumental and intrinsic/extrinsic—two items for each).
If it were considered desirable to increase reliability, it would be pos-
sible to include more items for each motivational type. If a more
dynamic view were required, it would be possible to administer the
questionnaire on more than one occasion (e.g. at the beginning and
the end of the semester) and to check how much motivation had
changed between the occasions.

Researchers interested in using any of the above need to be aware that

re-formatting might be required.

English Language Learning Strategy Inventory (ELLSI): 2016

Dear student: We are doing a research study about your language learn-
ing strategies. The research aims to test your current language learning
strategy level. The result of the questionnaire is only for research, it will
have no effect on your grades, and we will keep your personal informa-
tion confidential. You do not need to give your name or your email if you
do not wish to do so, though it might be useful if we want to contact you
about any of your answers. Since, however, we want to analyse the data
according to nationality and gender, we would appreciate you letting us
have that information. Thank you for your cooperation.
Name: ________________ email: ___________________
Nationality: _____________ Gender: M  F
Please read the following list of language learning strategies and mark each
one according to how often you use it. In addition, we would be interested to
have any comments you would like to make about the strategy items. Please
use the back of the sheet if you do not have enough room in the boxes
5=very often 4=often 3=sometimes 2=not often 1=never.
Item Strategy Rating Comment
1 Doing homework
2 Learning from the teacher
3 Reading books in English
4 Writing a diary in English
118  Developing Language Teacher Autonomy through Action Research

Item Strategy Rating Comment

5 Watching TV in English
6 Revising regularly
7 Listening to songs in English
8 Using language learning games
9 Writing letters in English
10 Listening to the radio in English
11 Talking to other students in English
12 Using a dictionary
13 Reading newspapers in English
14 Studying English grammar
15 Learning new vocabulary
16 Keeping a language learning notebook
17 Talking to native speakers of English
18 Learning by APPs on the mobile phone
19 Controlling schedules so that English study is
done
20 Taking online English courses
21 Not worrying about mistakes
22 Trying to think in English
23 Listening to native speakers of English
24 Learning from mistakes
25 Spending a lot of time studying English
26 Making friends with native speakers
27 Watching movies in English
28 Learning about the culture of English speakers
Any other strategies you are aware of using

I consent to these data being used for research and/or publication: (signature)

Language Skills Development (LSD): Strategy Questionnaire

Biodata: M/F Nationality___________

The following questionnaire contains some of the strategies which students
report using in order to assist the development of skills in the language they
are trying to learn. Please read the following strategy items and grade each one
according to the frequency with which you use it
1. very low 2. low 3. medium 4. high 5. very high
 5  Collecting the Data 
   119

Reading Skills
______1. I read extensively for information in the target language
______2. I read for pleasure in the target language
______3. I find reading material at my level
______4. I use a library to obtain reading material
______5. I first skim read a text then go back and read it more carefully
______6. I look for how a text is organized and pay attention to head-
ings and sub-headings
______7. I make summaries of what I read
______8. I make predictions about what I will read next
______9. I guess the approximate meaning by using clues from the
context
______10. I use a dictionary to get the exact meaning
Writing Skills
______1. I write letters or emails to friends in the target language
______2. When my mistakes are corrected, I learn from the corrections
______3. I write a variety of text types in the target language (e.g.
notes, messages, lists)
______4. I plan my writing before I start
______5. If I cannot think of the correct expression I think of another
way to express my meaning (e.g. synonyms)
______6. I use reference materials (e.g. a dictionary, thesaurus, or
grammar book) to check that what I am writing is correct
______7. If I am unsure about something I want to write I try to
express my meaning and do not worry too much about
correctness
______8. I write a rough copy before writing a good copy
______9. I write a diary in the target language
______10. I get someone to proofread my writing
Listening Skills
______1. I attend out-of-class events where I can listen to the new
language being spoken
______2. I use the media (e.g. radio, TV, or movies) to practise my
listening skills
120  Developing Language Teacher Autonomy through Action Research

______3. I listen to native speakers in public places (e.g. shops, res-

taurants, buses) and try to understand what they are saying
______4. I listen for key words which seem to carry most of the
meaning
______5. I predict what the other person will say next based on con-
text, background knowledge, or what has been said so far
______6. I ask the speaker to slow down, repeat, or clarify if I do not
understand
______7. I avoid translating what I hear word for word
______8. I use the speaker’s tone of voice, gestures, pauses, or body
language as a clue to meaning
______9. If I am unsure about meaning I guess in order to maintain
communication
______10. I listen carefully to how native speakers pronounce the lan-
guage I am trying to learn
Speaking Skills
______1. I repeat the new language to myself in order to practise it
______2. I seek out people with whom I can speak the target language
______3. I plan in advance what I want to say
______4. If I am corrected while speaking, I try to remember the cor-
rection and avoid making the same mistake again
______5. I ask questions in order to create conversation
______6. I do not worry about correctness as long as I can communi-
cate my meaning
______7. If necessary, I use gestures to convey my meaning and keep
a conversation going
______8. I practise the target language with other students
______9. If I do not know the vocabulary I want to use, I use similar
words or phrases or make them up
______10. I try to pronounce the target language like native speakers
Are there any other strategies which you have found useful for
developing language skills?
I consent to these data being used for research and/or
publication: (signature)
 5  Collecting the Data 
   121

Writing Strategy Survey (WSS)

Name:________ Class:________ Number:_______

Gender:_______ M   F Nationality:
The following is a list of strategies which some students report using
in order to develop their writing skills. Please read the strategies and rate
them according to the frequency with which you use them.
5=very high 4=high 3=medium 2=low 1=very low.
Rate Strategy
1 I learn as much new vocabulary as I can
2 I learn from my mistakes
3 I learn grammar rules
4 I use language I hear in my writing
5 I plan my writing
6 I use language I read in my writing
7 I write as much as I can in English (e.g. letters, emails, lists,
notes)
8 I revise new language regularly
9 I record new language I hear or read in a notebook
10 I memorize good texts as a model for my own writing
11 I write a diary in English
12 I check my writing for accuracy
13 I analyse texts (e.g. sentence structure, grammar, or figures of
speech)
14 I write creatively in English (e.g. stories or poetry)
15 I use references (e.g. dictionary or grammar book) to check my
writing
16 I write a rough copy then rewrite
17 I get someone to proofread what I have written
18 I learn idioms and other useful expressions
19 I notice the writing of native speakers
20 I consult my teacher if I have problems with my writing
Are there any other strategies you use to help you to improve your writing
skills?

I consent to the use of these data for research or publication

purposes: (signature)
122  Developing Language Teacher Autonomy through Action Research

Inventory of Language Learning Styles (ILLS)

Please rate each of the following learning style preferences according to
the scale:
5 = strongly agree; 4 = agree; 3 = neutral; 2 = disagree;
1 = strongly disagree.
I like to learn language Rating Comment
1 by reading
2 by writing things down
3 by speaking in the target language
4 by hearing the target language spoken
5 by seeing, e.g., diagrams, pictures
6 by moving around
7 by manipulating, e.g., models, cards
8 by learning the rules
9 by being corrected
10 with others
11 in an environment that I find pleasant
12 by memorizing
13 by having what I need to learn clear and
unambiguous
14 by concentrating on details
15 by thinking before speaking or writing
16 in order
17 by playing games
18 with authentic materials
Are there any other ways you like to learn language?

I consent to these data being used for research and/or

publication: (signature)

English Language Motivation Questionnaire (ELMQ)

Dear student: Please read the following statements and rate them accord-
ing to whether you.
5 = strongly agree; 4 = agree; 3 = neutral; 2 = disagree;
1 = strongly disagree.
Statement Rating Comments
1. I learn English so that I can improve myself
2. I learn English for my own satisfaction
3. My parents want me to study
   
5  Collecting the Data  123

Statement Rating Comments

4. My school wants me to study
5. I am interested in English culture
6. I want foreign friends
7. I think English is beneficial for me to study
8. I need to study abroad
Do you have any other kinds of motivation?

I consent to the use of this data for research or publication

purposes: (signature)

Interviews

Interviews are used to collect qualitative data sets through asking ques-
tions of the interviewee. Interviews can be done in three formats:
unstructured, semi-structured, and structured interviews, each of which
has different purposes and different advantages and disadvantages. For
example, structured interviews ensure that the interviewees are asked the
same questions in the same order. On the other hand, semi-structured
interviews are relatively more flexible in terms of the order and content of
the questions, which is quite applicable in qualitative research. This form
of interview gives researchers the opportunity to ask follow-up questions
to elicit more information if needed, although care should be taken not to
lead interviewees to provide particular answers. Unstructured interviews,
in which the interviewer does not ask pre-determined questions, but the
questions arise spontaneously, are the most flexible form of interview.
The interviewers might need to ask follow-up clarifying and elaborative
questions to elicit more in-depth responses.
If the interview is about learners’ vocabulary learning strategies, the
interview questions might be as given in Table 5.2 (based on Seidman
(2006)):

Format of Interviews

Interviews can be conducted individually or in groups. There are advan-

tages of interviewing learners one-on-one, such as individuals feel secure
124  Developing Language Teacher Autonomy through Action Research

Table 5.2  Interview stages with sample questions

Stages Sample questions
One Context of Can you tell me how your teacher teaches you
experience vocabulary in your classroom?
Two Details of Can you give specific examples of how you learn
experience vocabulary?
Three Reflection on How does teachers’ vocabulary teaching and your
meaning own strategies influence your vocabulary learning?

Table 5.3  Question types with examples

Question types Sample questions
Behaviour/experience What happens and what do you do in the
lesson?
Opinion/value What are the strengths and weaknesses of the
lesson?
Feeling How do you feel during the lesson?
Knowledge How is the lesson organized and conducted?
Sensory What changes do you see in your teaching?
Demographic/background Can you tell us about your background as a
student?

to say anything they want. On the other hand, group interviews create an
environment where each individual in the group could add to what oth-
ers are saying and a wealth of responses might be generated.

Content of Interviews

One of the critical aspects of interviews is the questions asked. An inter-

viewer needs to pay attention to the focus of each question. Patton (1987)
lists the types of questions which could address different dimensions of
data to be collected. Table 5.3 displays the types of questions with some
sample questions.

Observation

Observation is another way of collecting qualitative data especially about

classroom practices, student behaviours, and course stages. Observing
can be done in different formats by using different tools depending on
 5  Collecting the Data 
   125

Table 5.4  Sample observation questions

Questions Samples
Who How many students?
Where are they from?
How old are they?
What are the individual and group identities?
What What are the students doing?
Are there repetitive behaviours or irregular ones?
What are the resources used in activities?
How are the activities organized?
What is the nature of student interaction?
Who talks?
Who listens?
What are the student behaviours you can observe?
Where What is the physical setting like sights, sounds, smells, tastes, and
feelings?
When When do the students and teachers talk?
How long do the teacher and students talk?
What are the major ways of interacting with one another?
How What is the interrelationship of event and activities?
Are there any norms and rules that can be observed? If yes,
what are they?
Why What meanings can be attributed to activities and events?
What are the major reasons for suing such activities?

the research purpose and the questions. It is necessary to go to the class-

room with a general guide. The guide in Table 5.4 adapted form Lynch
(1996, p.  109) displays the specific questions that could be answered
­during or after the observation. Depending on the purpose for observa-
tion, the questions can be selected and prepared for use during the class.

Types of Observation

Observations in the classroom can be done in two major ways. Like dif-
ferent types of interviews, it is possible to go to the classroom with a pre-­
determined focus of observation or without.
Structured observations: Structured observations are carried out with
checklists and schedules which have pre-set objectives. The observer pays
attention to the items listed and rates them on the basis of the scale.
The following tools can be used as structured observation schemes in a
classroom.
126  Developing Language Teacher Autonomy through Action Research

• The Target Language Observation Scheme (TALOS) by Ullman and

Geva (1981)

(a) Low-Inference TALOS

(b) High-Inference TALOS

• The Communicative Orientation of Language Teaching (COLT)

observation scheme by Allen et al. (1984)

(a) for real-time coding with mix of low- and high-inference items
(b) from tape recording for student and teacher verbal interaction
with a time sampling

• Classroom observation checklist by Nunan (1989, pp. 147–148)

–– a series of statements are categorized against what happens in the
classroom
• Classroom observation tally sheet by Nunan (1989, p. 78)
• Structured observation schedule by Myhill (2002, p. 51)

Unstructured observations: For unstructured observations, forms may

not be necessary, although some kind of form may provide some guid-
ance for the observer. Such forms include open-ended note-taking spaces
on general sections or dimensions of the lesson. Observers are granted
the freedom to note down and describe what they observe in clear ways.
Some forms that can be used for this purpose are given below:

• The Reading for Science and Technology (REST) observation form by

Lynch (1996, pp. 71–72). The form includes:
–– Content/lesson events
–– Comments/observations
–– Reactions from teacher/researchers
Semi-structured observations: These observation forms allow for collect-
ing both structured data and unstructured data at the same time. While
there are pre-set categories for observation, there are also spaces for unex-
pected but relevant events and practices. One such form is:
 5  Collecting the Data 
   127

• University of California, Los Angeles (UCLA), observation form by

Brinton et al. (n.d.)

Questions to Consider
1. Some people suggest that ethical requirements have got to be so restric-
tive that it is almost impossible to do research any more, and we should
reduce the ethical barriers. What is your reaction to this idea?
2. Why might ethical issues be especially important in your context when it
comes to data collection?
3. What are the advantages/disadvantages of using pre-existing question-
naires rather than constructing your own?
4. Do you think any of the interview schedules suggested above might be
useful for the study you have in mind?
5. Would any of the observation forms be useful? Might you need to adapt
them, and, if so, how?

Example: An Action Research Study on Classroom Culture: Data

Collection

Permission for the research study was obtained from the university
authorities, and the questionnaire was handed out in class time, following
an explanation of the purpose of the study in order to ensure informed
consent. The students were asked to sign an agreement to having the data
used for research or publication purposes, and the survey was handed in
when students were finished. There were 38 questionnaires handed in,
but three were discarded because of incomplete data. (See the example
study in Chap. 3 or the Appendix for the questionnaire.)

Task
Specify how you will ensure your study is conducted ethically? What require-
ments are there to ensure ethical integrity in your context?
Specify exactly how you are going to collect the data for your study.
 6
Analysing the Data

Let us suppose that some interesting-looking data has been collected, and
now we want to know what to do with it. Let us be quite clear that we
do not actually have to do anything with it unless we want to. Maybe we
are quite happy just to have performed the collection exercise, perhaps we
have gained some insight into whatever it was that was puzzling us, and
that is as far as we want to go.
If, however, we want to go beyond this point, analysing the data can
be quite an intimidating prospect. So, let us have a look at it, and see if
we can break it down into manageable pieces which will produce robust
results.

Types of Data: Quantitative

Since there is often confusion with these basic concepts, and they are
not always easy for non-mathematicians to understand, the most impor-
tant from an action research point of view will be explained here. This
is important, since inappropriate analysis renders what might otherwise
have been interesting findings quite invalid.

© The Author(s) 2017 129

K. Dikilitaş, C. Griffiths, Developing Language Teacher Autonomy
through Action Research, DOI 10.1007/978-3-319-50739-2_6
130  Developing Language Teacher Autonomy through Action Research

1. Numerical data. This term refers to data which are actually real num-
bers. Examples are:

(a) Age—we can assume that someone who is 20 is twice as old as

someone who is 10.
(b) Income—someone earning $15,000 is earning half as much as
someone earning $30,000.
(c) Test scores—a student who scores 90 has scored 50% more than
a student who scores 60.
This kind of data is sometimes also called interval (because the val-
ues have regular intervals between them) or continuous (operating over
a range). Numerical data can be analysed using parametric tests (which
operate within set parameters, e.g. means, Pearson’s correlation, t-tests,
ANalysis Of VAriance (ANOVAs), Multivariate ANalysis Of VAriance
(MANOVAs), multiple regression) as long as they are normally distrib-
uted (see later section).
2. Ordinal data. This term is used to refer to data where particular atti-
tudes (such as level of agreement or disagreement) or subjective assess-
ment of a particular phenomenon (e.g. degree of frequency, level of
importance) are given numbers for the sake of convenience in that this
allows the data to be analysed by computer. Clearly, however, these fig-
ures are not really numbers (they are non-numerical). For example:

(a) We cannot say that someone who gives a questionnaire item a rat-
ing of 4 for agreement is twice as much in agreement as someone
who gives it a 2. In other words, we cannot assume that the inter-
val between the ratings is equal.
(b) It is clearly nonsense to say that the average of agree (4) and disagree
(2) is agree-and-a-half! In other words, this is not a continuous scale.

Likert-type instruments produce non-numerical data, and they should be

analysed using nonparametric tests such as medians, Spearman’s correlation,
and Mann–Whitney U or Kruskal–Wallis H tests of difference. Actually,
these tests are just as easy to use, and often produce results that are similar
to their parametric equivalents, though often slightly weaker. It is therefore
difficult to see why researchers should not use the appropriate tests.
 6  Analysing the Data 
   131

3. Nominal data. These kinds of data are even further removed from
“real” numbers than ordinal data. They are generated when variables
are divided into categories (hence they are also often called categorical
data) and given numbers, often quite arbitrarily, for the sake of being
able to enter them into a computer programme and analyse them sta-
tistically. For example:
(a) Males = 1, females = 2
(b) Chinese = 1, Europeans = 2, Africans = 3, Americans = 4
Clearly this kind of data is incapable of being analysed numerically.
How could we make any sense of it? If we add males (=1) and females
(=2) together, do we get 3? Would the average of a European and an
American in example (b) above be an African? When considered logically,
this is clearly nonsense. Nominal data can, however, be useful for group-
ing purposes, for instance if we want to find out if our male students
are doing better than our female students, or if students from a particu-
lar national background are doing better or worse than students from
elsewhere. Nonparametric tests of difference (e.g. Mann–Whitney U or
Kruskal–Wallis H) or chi-square (x2) should be used for this.

How Can I Do These Analyses?

Again, it is up to us. We can get some quite interesting results just
using a simple calculator. If we don’t have too many participants (and
this is typical of action research), it is easy enough to add the scores
together and divide by the number of scores, and there we have the
mean (average).
For those who know how to use Excel, there is a lot more available.
And these days there are a number of extremely user-friendly online plat-
forms such as Google forms, which can easily produce some very interest-
ing results.
But by far the best is SPSS (Statistical Package for Social Sciences).
For those who know how to use Excel, SPSS is not difficult—it operates
along fairly similar lines, but does so much more from a researcher’s point
of view. For those who want to get serious about research, SPSS is worth
the investment of a bit of time to learn how to use it.
132  Developing Language Teacher Autonomy through Action Research

Types of Statistical Procedures

In fact, there are many possible procedures, far too many for us to deal
with in depth here. Furthermore, the more complicated procedures often
depend on a knowledge of maths that the ordinary language teacher
rarely has. But let us look at the most common ones an action researcher
is likely to want to deal with:

1. Reliability
2. Factor analysis
3. Normality of distribution
4. Mean
5. Frequencies
6. Percentages
7. Median
8. Mode
9. Correlation
10. Difference
11. Effect size

We will use the small-scale action research study on classroom culture

we have been using in previous chapters to provide examples which we
will work through here together.

• In the case of this study, there were 35 survey sheets (actually, not so
many!), each with nine items, each of which has a rating from 1 to 5.
• In addition, since we want to be able to see if female students perceive
classroom culture differently from their male classmates, we need to go
through and code each one according to gender (male = 1, female = 2).
• Furthermore, we are also interested to know if the local students view
classroom culture differently from international students, so this is also
coded (local = 1, international = 2).
• All of this has to be entered onto the SPSS spreadsheet, under the vari-
able headings GEN (gender), NAT (nationality) and Q1–9 for the
questionnaire items.
 6  Analysing the Data 
   133

GEN NAT Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9

2.00 1.00 4.00 4.00 5.00 4.00 4.00 5.00 4.00 5.00 4.00

1.00 2.00 4.00 5.00 3.00 2.00 5.00 4.00 1.00 2.00 3.00

1.00 2.00 4.00 2.00 3.00 4.00 3.00 1.00 2.00 4.00 1.00

1.00 2.00 2.00 2.00 4.00 5.00 4.00 1.00 4.00 5.00 1.00

2.00 1.00 2.00 4.00 4.00 5.00 4.00 4.00 4.00 5.00 4.00

1.00 1.00 2.00 2.00 4.00 1.00 2.00 2.00 3.00 3.00 2.00

2.00 1.00 3.00 4.00 4.00 3.00 4.00 2.00 4.00 5.00 4.00

2.00 1.00 4.00 4.00 4.00 3.00 2.00 3.00 5.00 5.00 3.00

1.00 2.00 4.00 3.00 4.00 4.00 4.00 3.00 5.00 5.00 4.00

2.00 2.00 4.00 4.00 5.00 3.00 3.00 2.00 5.00 5.00 3.00

Fig. 6.1  Sample SPSS spreadsheet

When this is done, we will have a grid, the first ten rows of which will
look a bit like Fig. 6.1:
Having got this far, we can now begin analysing in earnest! For the sake
of exemplification, we will work through procedures 1–11 listed above.

Reliability

Reliability refers to the extent to which the items in our instrument are
consistent. In order for our instrument to be considered a valid measure
of the concept we are trying to measure, it must be reliable. In order to
analyse for reliability:

• Go up to the task bar at the top, click ANALYZE, go down to SCALE,

go across to RELIABILITY ANALYSIS
• Highlight the items you want to test (in this case Q1–9) and click
them across into the treatment box
• Click OK
134  Developing Language Teacher Autonomy through Action Research

Reliability Statistics

Cronbach's N of
Alpha Items

.776 9

Fig. 6.2  Sample reliability statistics

A new screen will come up including the box in Fig. 6.2:

This tells us that the Alpha reliability is 0.776. This is not a high level
of reliability (in the 0.9 range is best), but for such a relatively small
number of participants (N = 35) it is not bad, and anything above 0.7
(or sometimes even in the 0.6 range) is usually considered to be OK (e.g.
Dörnyei 2007). The chances are that if we carried on and gathered more
data from more participants, reliability levels would rise, since, when
numbers are small “outliers” (i.e. participants whose response is consider-
ably different from the majority) have more of an effect than if it were
part of a much larger sample.
If, however, we would really like to get a higher level of reliability from
this study, we might wonder if there may be one item or a group of items
bringing the reliability co-efficient down. In order to check if this is the
case:

• Go back into the main matrix, re-click ANALYZE, go down to

SCALE, across to RELIABILITY ANALYSIS
• Before you click OK, click STATISTICS, then SCALE IF ITEM
DELETED
• Click CONTINUE, then OK

The screen will include the grid in Fig. 6.3:

From the far right-hand column of this grid we can see that there is
no item which, if it were deleted, would raise the reliability substantially,
although if we removed Q6 (about eating and drinking in class), the reli-
ability would go up to 0.791, which would round to 0.80, which actually
sounds much better. So, is this worth doing or not? Perhaps if we try a
factor analysis, we might get some more information to help us make up
our minds.
 6  Analysing the Data 
   135

Item-Total Statistics
Scale Mean Scale Corrected Cronbach's
if Item Variance if Item-Total Alpha if Item
Deleted Item Deleted Correlation Deleted
Q1 28.0000 31.412 .549 .744
Q2 27.5143 30.257 .553 .740
Q3 27.1143 30.457 .616 .734
Q4 28.1714 32.440 .357 .769
Q5 27.8571 30.303 .473 .752
Q6 28.3714 34.534 .198 .791
Q7 27.0000 28.765 .607 .730
Q8 26.7714 29.829 .531 .743
Q9 28.1143 33.575 .288 .778

Fig. 6.3  Grid for reliability if item deleted

Factor Analysis

A factor analysis is designed to explore whether certain items “hang

together”, and they can be daunting both to conduct and later to inter-
pret. And there are many different ways they can be done in order to
produce different results. But a simple factor analysis does not have to be
so complicated, so let us see how we can perform a factor analysis on the
data obtained for our classroom culture survey.

• Click ANALYZE, down to DIMENSION REDUCTION, across to

FACTOR
• Transfer items Q1–9 to the variable box
• Click OK

The next screen includes the grid in Fig. 6.4:

From this we can see that the programme’s default settings have
divided our questionnaire into three factors. If we look for the high-
est loadings (the largest numbers), we can see that by far the majority
load most highly onto factor 1. Of the rest, there is only one (Q6)
which loads most highly onto factor 2, and only two (Qs 4 & 9)
136  Developing Language Teacher Autonomy through Action Research

Component Matrixa
Component
1 2 3
Q1 .691 .379 -.256
Q2 .670 .538 .216
Q3 .767 -.150 -.350
Q4 .472 -.484 .528
Q5 .617 .313 .237
Q6 .297 .449 -.375
Q7 .765 -.352 -.151
Q8 .695 -.597 -.149
Q9 .369 .213 .639

Extraction method: Principal Component Analysis.

a. 3 components extracted

Fig. 6.4  Sample factor analysis grid (three-factor solution)

which load most highly onto factor 3. These do not seem to provide
us with anything very useful by way of grouping the items into any-
thing at all thematically coherent. Indeed, we always saw this small
questionnaire as a coherent whole in itself, designed to measure ideas
of polite classroom behaviour, and we never really intended to divide
it. So perhaps we should see what we get if we try for a one-factor
solution:

• Click ANALYZE, down to DIMENSION REDUCTION, across to

FACTOR
• Transfer items Q1–9 to the variable box
• Before you click OK, click EXTRACTION, then, in the EXTRACT
box, click FIXED NUMBER OF FACTORS, and put 1 in the box
• CONTINUE, OK

You will get a page which includes the grid in Fig. 6.5

From this we can see that all of the items do, in fact, load onto one fac-
tor, although the loading for our slightly troublesome Q6 is the lowest.
Even this, however, is only slightly under 0.3 (often considered a cut-off
 6  Analysing the Data 
   137

Component Matrixa
Component
1
Q1 .691
Q2 .670
Q3 .767
Q4 .472
Q5 .617
Q6 .297
Q7 .765
Q8 .695
Q9 .369

Extraction method: Principal Component Analysis

a. 1 component extracted

Fig. 6.5  Sample one-factor solution grid

point), and it would actually round to 0.30 if we chose to work with two
places of decimals.
We are therefore left with the decision as to whether to leave Q6 in
place, and live with slightly lower reliability, or to remove it (the machine
cannot make all our decisions for us). In fact, as the researcher, I had
found myself quite interested in some of the responses to Q6 about eat-
ing or drinking in class, so I felt that this outweighed the possible threat
to reliability posed by a slightly low-loading item.

Normality of Distribution

Another test which should be performed before proceeding with further

tests (e.g. tests of central tendency such as mean, median, mode, or cor-
relation/difference tests) is for normality of distribution. This refers to the
well-known bell curve, where means are more or less evenly distributed
around a central axis. Parametric tests (e.g. Pearson’s product–moment
correlation, t-tests, ANOVAs) operate on an assumption of normality.
If, therefore, the data are not normally distributed, nonparametric tests,
138  Developing Language Teacher Autonomy through Action Research

Tests of Normality
Kolmogorov-Smirnova Shapiro-Wilk
Statistic df Sig. Statistic df Sig.
Q1 .271 35 .000 .803 35 .000
Q2 .265 35 .000 .871 35 .001
Q3 .271 35 .000 .793 35 .000
Q4 .205 35 .001 .913 35 .009
Q5 .207 35 .001 .904 35 .005
Q6 .199 35 .001 .913 35 .009
Q7 .303 35 .000 .723 35 .000
Q8 .417 35 .000 .593 35 .000
Q9 .243 35 .000 .889 35 .002

a. Lilliefors Significance Correction

Fig. 6.6  Sample tests for normality of distribution

which do not assume normality, should be used. The normality of distri-

bution can be determined as follows:

• Click ANALYZE, then DESCRIPTIVES, then EXPLORE

• Move the variables you want to test (in this case, Q1–9) to the depen-
dent list
• Click PLOTS, then NORMALITY PLOTS WITH TESTS
• CONTINUE, OK

A grid that is something like in Fig. 6.6 will appear among others on

the page.
If the Sig. figure for these tests is less than 0.05, the distribution is not
normal. As we can see, the significance figures for both of these tests for
all items are less than 0.05. In other words, none of the distributions are
normal, meaning that we need to use nonparametric tests of correlation
and difference. As we can see, the results for both tests are similar, but, if
we need to choose one or the other, the Kolmogorov–Smirnov is usually
considered the more appropriate one for smaller samples (less than 100).
In fact, this is a very usual result for ordinal data, meaning that non-
parametric tests are by far the most commonly appropriate ones.
 6  Analysing the Data 
   139

Mean

Statistics such as the mean, median, mode, frequencies, and percentages

are called descriptive statistics, since they are used to describe the results
obtained from a study of a particular sample. The mean actually applies
to numerical data which are normally distributed. Since we have ordinal
data, and we have already found the distribution is not normal, means are
not appropriate for this study. We could, perhaps, have been interested in
a mean if we had included test scores (which are numerical) in our data
collection (and we had found them to be normally distributed). Since
this is actually a very common thing to do, for the record:

• Go to ANALYZE, go down to DESCRIPTIVE STATISTICS, go

across to DESCRIPTIVES
• Transfer the item/s you want to analyse to the VARIABLE box
• Click OK

The programme will produce a box with the ITEMS, the MINIMUM
and MAXIMUM values, the MEAN, and the STANDARD DEVIATION
(which indicates the average distance of the scores from the mean).

Frequencies

If we want to know how often a particular value occurs, SPSS will calcu-
late frequencies for us:

• Go to ANALYZE, go down to DESCRIPTIVE STATISTICS, go

across to FREQUENCIES
• Move target items (in this case, Q1–9) to VARIABLE box
• Click OK

A page will appear which includes a matrix for each of the items
(Fig. 6.7).
From the “Frequency” column of this matrix we can see that for Item
1, there were 3 students who gave a rating of 1, 6 with a rating of 2, 10
140  Developing Language Teacher Autonomy through Action Research

Q1
Frequency Percent Valid Cumulative
Percent Percent
1.00 3 8.6 8.6 8.6
2.00 6 17.1 17.1 25.7
Valid 3.00 10 28.6 28.6 54.3
4.00 16 45.7 45.7 100.0
Total 35 100.0 100.0

Fig. 6.7  Sample frequency matrix

with a rating of 3, 16 with a rating of 4, and 0 with a rating of 5, adding

up to a total of 35 (the number of our participants). The SPSS page will
give similar matrices for each item.

Percentages

The same matrix gives us percentages. From the “Percent” column, we

can see that 8.6% of the participants gave a rating of 1, 17.1% a rating of
2, 28.6% a rating of 3, 45.7% a rating of 4, adding up to a total of 100%.
All of the percentages are valid, meaning we do not have to go look-
ing for problems with the data entry or elsewhere, and the “Cumulative
Percent” simply progressively adds the percentages to 100%.

Median

The median indicates the midpoint of a set of numbers, and is appropri-

ate for use with non-numerical data and/or data which are not normally
distributed. In order to calculate the median:

• Go to ANALYZE, down to DESCRIPTIVE STATISTICS, across to

FREQUENCIES
• Transfer target items to VARIABLE BOX
• Click STATISTICS, then MEDIAN
• Then CONTINUE, then OK
 6  Analysing the Data 
   141

Statistics
Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9
Valid 35 35 35 35 35 35 35 35 35
Missing 0 0 0 0 0 0 0 0 0
3.00 4.000 4.000 3.000 3.000 3.000 5.000 5.0000 3.0000
Median

Fig. 6.8  Sample median matrix

A page will appear which includes a matrix like the one given in
Fig. 6.8
On the last line, this matrix gives us the medians for each item. We
can see that the items with the highest medians are Q7 and Q8 (using
inappropriate terms of address, and using bad language); in other words,
these are the items of which most students disapproved most strongly.
Items 2 and 3 (using a phone and talking while someone else is talking)
received medians of 4, indicating a high level of perceived impoliteness.
The remainder of the items received medians of 3, suggesting that, over-
all, these did not arouse strong opinions among the students. None of the
medians, however, were less than 3, indicating that none of the surveyed
behaviours were considered actually polite.

Mode

The mode indicates the most frequent “popular” or “fashionable” rating.

To get this,

• Go to ANALYZE, down to DESCRIPTIVE STATISTICS, across to

FREQUENCIES
• Transfer target items to VARIABLE BOX
• Click STATISTICS, then MODE
• Then CONTINUE, then OK

A page will appear which includes the matrix provided in Fig. 6.9.

142  Developing Language Teacher Autonomy through Action Research

Statistics
Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9
Valid 35 35 35 35 35 35 35 35 35
Missing 0 0 0 0 0 0 0 0 0
4.00 4.00 4.00 3.00 4.00 2.00 5.00 5.00 3.00
Mode

Fig. 6.9  Sample mode matrix

From this we can see, that although Item 1 received a median of 3, the
most common rating was actually 4, suggesting that a large number of
the students consider coming late to be impolite. Interestingly also, the
mode for Item 6 (about eating or drinking in class) was 2, indicating that
many students actually do not have too much of a problem with this,
although the median for this item was 3.

Correlations

Correlations are a way of finding the relationship between variables,

from which we can infer relationships among a wider population (hence
these kinds of procedures are called inferential statistics). There are dif-
ferent tests of correlation which might be used. The Pearson product–
moment test is used for numerical data, such as exam scores and ages
(assuming these are normally distributed). For non-numerical data
(such as from Likert-­type questionnaires) a nonparametric test (such as
Spearman’s rank order correlation) should be used. Correlations can be
used with numerical or ordinal data, but not with nominal data (such
as gender or nationality).
In the case of the small-scale exploration of student perceptions of
classroom politeness we have been using to illustrate research procedures
in this volume, there was no exam score available at the time of the study
which might have been used to correlate exam success with perceptions
of politeness (possibly quite an interesting question!). Furthermore, since
the questionnaire was completed anonymously in order to promote hon-
est rating, it would have been impossible to match the ratings and the
scores, even if they had been available. So the best we can do in the case
 6  Analysing the Data 
   143

of this study is correlate the items with each other, which might tell us,
for instance, whether students who disapprove of coming late also disap-
prove of various other behaviours. In order to do this:

• Go to ANALYZE, then down to correlate, than across to bivariate

• Transfer target items (Q1–9) to VARIABLE box
• Unclick PEARSON (the default) and click SPEARMAN
• Click OK

The matrix in Fig. 6.10 will be produced (note that, except for col-
umn 1, reproduced in full for the sake of exemplification, only significant
results have been reproduced here because of space constraints).
So, what does it mean? First of all, it is necessary to understand that the
asterisks (e.g. .367*, .493**) indicate probability, which is the likelihood
that the relationship between the two variables in question is more than
would be expected merely by chance. In the first case, .367* indicates
that this probability is at the p < .05 level; in other words, there is more
than a 95% likelihood (also referred to as the confidence interval) that
the relationship is more than chance. In the second case, .493** indicates
a probability rate of p < .01, or a 99% likelihood that the relationship is
more than chance.
Note also that a correlation coefficient of 1.00 goes diagonally from
top left to bottom right. This is because these boxes are correlating items
with the same item; therefore the correlation is perfect (=1).
From the matrix we can see that there are significant relationships at
the p < .01 (99% probability) level between

• Q1 and Q2. In other words, there is a strong probability that students

who think that coming late is impolite will also think that using a
phone in class is impolite.
• Q2 and Q5. In other words, there is a strong probability that students
who think that using a phone in class is impolite will also think that
sleeping in class is impolite.
• Q7 and Q8. In other words, students who think that it is impolite to
use inappropriate terms of address will think the same about the use of
bad language.
144  Developing Language Teacher Autonomy through Action Research

Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9
Q1 Correlation 1.00 .493**
coefficient
- .003
Sig (2-tailed)
35 35
N
Q2 Correlation .493** 1.00 .538** .367*
coefficient
.003 .001 .030
Sig (2-tailed)
35 35 35
N
Q3 Correlation .325 1.00
coefficient
.057
Sig (2-tailed)
35
N
Q4 Correlation .028 1.00
coefficient
.875
Sig (2-tailed)
35
N
Q5 Correlation 321 .538** 1.00
coefficient
.060 .001
Sig (2-tailed)
35 35
N
Q6 Correlation .332 1.00
coefficient
.051
Sig (2-tailed)
35
N
Q7 Correlation 328 1.00 .604**
coefficient
.054 .000
Sig (2-tailed)
35 35
N
Q8 Correlation .134 .604** 1.00
coefficient
.443 .000
Sig (2-tailed)
35 35
N
Q9 Correlation .171 .367** 1.00
coefficient
.326 .030
Sig (2-tailed)
35 35
N

**. Correlation is significant at the 0.01 level (2-tailed).

*. Correlation is significant at the 0.05 level (2-tailed).

Fig. 6.10  Sample Spearman’s correlation matrix

 6  Analysing the Data 
   145

We can also see that there is a relationship at the p < .05 level between
Q2 and Q9. In other words, there is a 95% likelihood that the relation-
ship between perceptions of using a phone in class and wearing inappro-
priate clothing is more than just chance.
So we have established that there are significant relationships between
some of the variables, but it is now the researcher’s job to decide whether
these relationships are at all important, and, therefore, whether they are
worth reporting and discussing. Is it at all important for us to know that
Q1 is related to Q2, or Q7 to Q8? What use is this likely to be to any-
body? This is a decision that only the researcher can make based on the
aim of the study and the intended audience.

Differences

Differences are another kind of inferential statistic which can be used

to generalize beyond the immediate sample. Nominal categories, such
as gender or nationality, where arbitrary numbers are assigned to par-
ticular groups for the sake of convenience for entering them into a com-
puter programme, cannot be analysed by means of correlations. But such
­categories can be used as a grouping mechanism for the purpose of exam-
ining differences.
In the case of our sample study of politeness perceptions in the lan-
guage classroom, there are two variables (gender and nationality) which
might provide interesting insights into different expectations of polite
behaviour. Since we have already found that the data from this study are
not normally distributed, we cannot use t-tests, so we will choose non-
parametric tests. If we look first of all at gender:

• Click ANALYZE, go down to NONPARAMETRIC TESTS, go across

to LEGACY DIALOGUES, then 2 INDEPENDENT SAMPLES
• Move target variables (Q1–9) across to the TEST VARIABLE LIST
box
• Move “gender” into the GROUPING VARIABLE box
• Click DEFINE GROUPS, and put 1 (male) in the first box and 2
(female) in the second
146  Developing Language Teacher Autonomy through Action Research

Test Statisticsa
Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9
Mann- 139.00 92.500 139.50 142.50 117.00 94.000 102.00 128.50 116.00
Whitney 0 0 0 0 0 0 0
U
Wilcoxo 392.00 183.50 230.50 233.50 208.00 347.00 193.00 219.50 207.00
nW 0 0 0 0 0 0 0 0 0
Z -.146 -1.803 -.128 -.018 -.913 -1.728 -1.540 -.622 -.967
Asymp. .884 .071 .898 .986 .362 .084 .124 .534 .334
Sig. (2-
tailed)
Exact .906b .085b .906b .987b .389b .098b .169b .625b .371b
Sig.
[2*(1-
tailed
Sig.)]

a. Grouping Variable: Gender

Fig. 6.11  Sample matrix for Mann–Whitney U test of difference—gender

• Choose MANN–WHITNEY U if it is not already selected

• Click CONTINUE, then OK

The page produced will include the matrix provided in Fig. 6.11.

Bearing in mind that a probability value of anything over 0.05  in
not usually considered significant, we can see from the Asymp. Sig line
that none of these differences is significant. In other words, although
there might, perhaps, be some differences according to gender (maybe
a higher or lower median), such differences as there might be are not
large enough to be considered more than might be expected merely by
chance.
But it should be remembered that this may, in itself, be an interesting
finding, since gender is often believed to affect perceptions of behaviour
such as politeness. From this point of view, to discover that there is actu-
ally no significant difference is as interesting as if we had found some-
thing significant.
 6  Analysing the Data 
   147

Test Statisticsa
Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9
Mann- 74.500 99.500 94.000 98.500 120.00 117.50 127.50 124.50 129.00
Whitney 0 0 0 0 0
U
Wilcoxo 374.50 399.50 394.00 398.50 420.00 417.50 193.50 190.50 195.00
nW 0 0 0 0 0 0 0 0 0
Z -2.183 -1.208 -1.442 -1.235 -.438 -.532 -.176 -.335 -.112
Asymp. .029 .227 .149 .217 .661 .595 .860 .738 .911
Sig. (2-
tailed)
Exact .040b .252b .186b .238b .687b .612b .875b .793b .930b
Sig.
[2*(1-
tailed
Sig.)]

a. Grouping Variable: Nationality

Fig. 6.12  Sample matrix for Mann–Whitney U test of difference—nationality

If we want to find out if there are any differences according to nation-

ality (Turkish and “others”), we follow the same procedure through to
the point of defining the grouping variable, when we select the “nation-
ality” rather than the “gender” variable. The matrix in Fig. 6.12 will be
produced:
In the case of this analysis, we can see that Item 1 (coming late to
class) shows a significant difference according to nationality (p = .029).
But which group is it that considers unpunctuality more impolite? If
we look at the matrix above the “Test Statistics” matrix, we find this
(Fig. 6.13).
From this matrix we can see that the mean rank for Item 1 for
Turkish students (group 1) is only 15.60, while the mean rank for
other nationalities (group 2) is 23.23. In other words, international
students consider unpunctuality considerably more impolite than
Turkish students.
148  Developing Language Teacher Autonomy through Action Research

Ranks
Nat N Mean Sum of
Rank Ranks
1.00 24 15.60 374.50
Q1 2.00 11 23.23 255.50
Total 35
1.00 24 16.65 399.50
Q2 2.00 11 20.95 230.50
Total 35
1.00 24 16.42 394.00
Q3 2.00 11 21.45 236.00
Total 35
1.00 24 16.60 398.50
Q4 2.00 11 21.05 231.50
Total 35
1.00 24 17.50 420.00
Q5 2.00 11 19.09 210.00
Total 35
1.00 24 17.40 417.50
Q6 2.00 11 19.32 212.50
Total 35
1.00 24 18.19 436.50
Q7 2.00 11 17.59 193.50
Total 35
1.00 24 18.31 439.50
Q8 2.00 11 17.32 190.50
Total 35
1.00 24 18.13 435.00
Q9 2.00 11 17.73 195.00
Total 35

Fig. 6.13  Sample mean ranks grid

This small-scale study identified only two groups according to nation-

ality, since to have divided the “other nationalities” according to their
actual origins (e.g. Greece, Turkmenistan) would have produced unviably
small groups and emphasized even more the numerical dominance of the
Turkish students. If, however, there were sufficient numbers to support
 6  Analysing the Data 
   149

analysis of more than two groups, this can be done with a Kruskal–Wallis
H test, which is, effectively, the nonparametric equivalent of an ANOVA
with parametric data. To do this:

• Follow the above procedure to LEGACY DIALOGUES, then go

across to K INDEPENDENT SAMPLES
• Move the target variables to the TEST VARIABLE LIST, and the
grouping variable to the GROUPING VARIABLE box
• In the DEFINE RANGE BOX, identify the MAXIMUM and the
MINIMUM values
• Then CONTINUE and OK

Something similar to the “Test Statistics” matrix above will be pro-

duced which will tell you whether there is a significant difference accord-
ing to the grouping variable (p less than .05). To determine which of the
groups rated the item more or less highly, refer to the “Ranks” matrix on
the same page.

Effect Size

Increasingly, journals are requiring reporting of effect size, since this is

not dependent on sample size as other statistical procedures are. For those
who aspire to publication, knowing how to produce effect sizes may be
necessary.
Confusingly, there seem to be numerous ways of calculating and inter-
preting effect sizes, but a common and relatively straightforward way is
to calculate the Eta value for parametric data:

• Go to ANALYZE, then down to DESCRIPTIVE STATISTICS, then

across to CROSS TABS
• Transfer the target dependent variables into the ROW/S box
• Transfer the target independent variable into the COLUMN/S box
• Click STATISTICS, then ETA
• Then CONTINUE, then OK
150  Developing Language Teacher Autonomy through Action Research

Table 6.1  Approximate effect size thresholds for Eta values

Size Eta value
Small .1
Medium .24
Large .38

The programme will produce Eta values for each of the dependent
variables.
And what does it mean? According to Cohen (1988), the strength of
an Eta value can be assessed according to the following “rule-of-thumb”
thresholds (Table 6.1).
When calculating effect size from a Mann-Whitney, divide the Z figure
by the square root of the number of participants (Yanati, K. http://yatani.
jp/teaching/doku.php?id=hcistats:mannwhitney#effect_size) When cal-
culating from a Kruskall-Wallis, you can use Wilson, D. Practical Meta-­
Analysis Effect Size calculator:  https://www.campbellcollaboration.org/
escalc/html/EffectSizeCalculator-R5.php.
It is important to bear in mind, however, that “the main short-
coming of effect size… [is] that there are no universally accepted and
­straightforward indices to describe it” (Dörnyei 2007, p. 212). Indeed,
the APA Publication Manual itself lists more than a dozen different ways
of estimating effect size.

Types of Data: Qualitative

Increasingly, qualitative research methods are gaining recognition for the
human dimension that they can add to quantitative methods. Perhaps
quantitative methods may be especially useful in a mixed methods
approach, where qualitative interpretations of data provided by par-
ticipants can add useful insights to otherwise de-personalized sets of
numbers.
There are two major ways of approaching a qualitative data set:
deductive and inductive. In the former, the researcher analyses against
 6  Analysing the Data 
   151

Table 6.2  Characteristics of deductive and inductive approaches to data analysis

Deductive Inductive
Follows a top-down strategy Follows a bottom-up strategy
Depends on external themes and Is data-driven
categories
Tests external themes and categories Induces themes and categories
Categorizes content mechanically Synthesizes content creatively
Is confirmatory and static Is exploratory and dynamic
Supports existing literature Interprets new meanings

pre-set categories or themes. This is usually a practice in studies in

which particular dimensions are designed in a close-ended question-
naire. The verbal responses are evaluated around these dimensions
without inducing any new category from the data. Such data analysis
practices are confirmatory in the sense that the researcher checks the
data and confirms whether the category is evident in the data. Themes
and categories are gathered from the existing relevant literature through
careful synthesis of different views, or the dimensions of the borrowed
questionnaires may already contain themes and categories that allow for
categorization of the data.
In an inductive approach to data analysis, the researcher induces
emerging themes from the data as suggested by grounded theory through
an exploratory perspective. The researcher starts the analysis with no prior
selection of categories and themes. Rather than test hypotheses or theo-
ries, this kind of data analysis is used to build potential theories about
the phenomenon under investigation. Inductive analysis requires careful
reading and critical thinking to bring out themes and build them into a
meaningfully related whole. The characteristics of the two approaches to
data analysis are displayed in Table 6.2.

Analysing Interview Responses

The data collected and recorded through interviews should be transcribed

carefully. However, rather than transcribing all the data, it is possible
152  Developing Language Teacher Autonomy through Action Research

to select just the necessary information from what the participants have
said. Three basic steps could be:

• Read these manuscripts again and again until no more meanings can
be drawn (saturation)
• Identify and underline words or groups of words that have negative or
positive meanings relating to the theme being researched (open coding)
• Seek meaning relations among all these words (emerging themes) and
look for thematic groupings (axial coding)
• Look for an overall theme which will unify the study thematically
(selective coding)

Analysis of Observational Data

Analysing observational data depends on how data were collected:

through structured (checklists or schemes) or unstructured (open-ended,
thick notes) methods.
Structured observations can be analysed through ranking, percentages,
mean scores of behaviours, interactions, or practices. Such analysis could
help researchers present a set of data quantified from qualitative data.
For example, Fig. 6.14 shows how many times the teacher used different
types of error correction strategies in three different lessons.

Error correcon types

10

0
Explicit Recast Metalingisc Elicitaon Clarificaon

Lesson 1 Lesson 2 Lesson 3

Fig. 6.14  Teacher’s error correction strategies

 6  Analysing the Data 
   153

This is an example of how qualitative data is quantified in a simple

manner and shown in a bar chart. The numbers indicate that the teachers’
error correction practices are usually explicitly made and involve metalin-
guistic error correction. According to the results of this observation, the
teacher does not frequently resort to implicit correction strategies such as
elicitation, clarification, and recast. This is at least the case with the three
lessons observed.
Unstructured observation is analysed through different ways. It depends
on how the data were collected: time-oriented or theme-­oriented. If it is
collected on the basis of the particular periods such as the first 10 min-
utes, the second 10 minutes, and so on, then whatever happened during
these particular intervals is grouped together. On the other hand, if it
is theme-based, then each theme is exemplified and described in detail
with reference to how it happened in the classroom. For example, if the
research question is

• How does the teacher correct students’ verbal mistakes?

a checklist containing types of error correction can be prepared and

ticked as the teacher practises each type. The frequency of error correc-
tion that happens is analysed quantitatively. In order to understand the
nature of the mistakes, the conversation and interaction pattern between
the teacher and the student might need to be written down; or a video-
tape of the lesson will help to identify the error correction and how cor-
rections are made.

Reliability of Qualitative Results

During the analysis of the qualitative data, there are particular strategies
to follow which could enhance the trustworthiness of the analysed data.
The findings from the analysis of verbal or written data are often criticized
for being too subjective in that they are drawn from the researcher’s own
perspectives. To minimize the risk of drawing subjective results, research-
ers need to consult a second view about the appropriateness of the catego-
ries and themes that they induced. If this is done, it is possible to report
inter-­rater reliability by calculating the number of categories identified by
154  Developing Language Teacher Autonomy through Action Research

one researcher, the number of categories identified by another researcher,

and expressing the relationship as a percentage. This practice is often
skipped or not reported properly, which makes readers suspicious of the
accuracy and authenticity of the data collection and analysis process.
Another way of safeguarding reliability is by means of member-­checking,
which refers to the process of getting approval from the research partic-
ipants for the data transcribed, analysed, and reported in the paper. At
each stage, it is necessary to work closely with the participants and share
with them the analysed data. The participants might approve or ask for
modifications or changes in the particular statements which do not express
properly what they wanted to say. This process adds to the results’ trustwor-
thiness, and needs to be explicitly documented in the data analysis section.
Debriefing is another way of developing the level of trustworthiness of
the data. This time colleagues or other researchers or co-authors of the
paper can be consulted to check the accuracy and appropriateness of the
categories and themes. While selecting such people, it is necessary to pay
attention to whether they know about the research topic and have experi-
ence in qualitative data coding. Otherwise, inappropriate confirmation
may result which could spoil the findings. The ultimate goal is to find out
the level of inter-coder reliability. Again, all this process should be
reported in the methodology section with a detailed discussion of the
confirmed and disconfirmed categories and themes.

Questions to Consider
  1. What is the difference between parametric and nonparametric data?
  2. What is the difference between numerical, ordinal, and nominal data?
  3. To what does reliability refer?
  4. What does a factor analysis tell us?
  5. If we say data is/not normally distributed, what do we mean?
  6. How are mean, median, and mode different?
  7. What do correlations tell us?
  8. What are two common nonparametric tests of difference?
  9. To what does the Eta value refer?
10. Why can qualitative data be useful?
11. What are the three coding stages?
12. What are some ways of safeguarding reliability?
   
6  Analysing the Data  155

Example: An Action Research Study on Classroom Culture: Data

Analysis
Quantitative data
Since the quantitative data obtained from the Likert-type questionnaire
used in this study are ordinal, the ratings were analysed for medians, as
shown in Table 6.3
We can see that the items with the highest medians are Q7 and Q8 (using
inappropriate terms of address and using bad language); in other words,
these are the items of which most students disapproved most strongly.
Items 2 and 3 (using a phone and talking while someone else is talking)
received medians of 4, indicating a high level of perceived impoliteness.
The remainder of the items received medians of 3, suggesting that, over-
all, these did not arouse strong opinions among the students. None of the
medians, however, were less than 3, indicating that none of the surveyed
behaviours were considered actually polite.
Qualitative data
In order to analyse the comments that students were asked to add to
explain their ratings, a grounded approach was adopted. This involved
overviewing the data and noting down all the themes presented (open-
coding stage). Next, these sometimes rather fragmented ideas were orga-
nized around a central theme (axial coding). Finally, an overarching theme
was identified, which unified all the ideas presented (selective coding). In

Table 6.3  Median ratings for impolite classroom behaviour

No. Item Median
1 Coming late 3
2 Using a phone 4
3 Talking while someone else is talking 4
4 Not paying attention 3
5 Sleeping 3
6 Eating or drinking 3
7 Using inappropriate terms of address 5
8 Using bad language 5
9 Wearing inappropriate clothing 3
156  Developing Language Teacher Autonomy through Action Research

order to check reliability, a member-checking procedure was adopted; in

other words, the results were taken back to the class and discussed, and
some changes/additions were made according to the students’ feedback
before the final coding classifications were decided.
Open coding
At this stage, the comments made by the students were reviewed (not all
students made comments about every item) and noted down under the
relevant questionnaire item:

1. Coming late

• Being on time is your responsibility

• Causes lack of concentration
• If it is temporary OK, but if they keep coming late it is rude

2. Using a phone

• It is disrespectful
• It disturbs others
• OK for emergency, but it causes students to lose their concentration

3. Talking while someone else is talking

• We should behave as we want to be treated

• Plain rude
• Everybody needs to respect one another

4. Not paying attention

• This is disrespectful
• It distracts others
• Don’t attend the class if you won’t pay attention anyway
 6  Analysing the Data 
   157

5. Sleeping

• There is always coffee

• It is disrespectful
• Bedrooms are for sleeping, not classes

6. Eating or drinking

• Sometimes we don’t get time for lunch or breakfast

• OK unless loud or smells bad
• Maybe he/she has an illness

7. Using inappropriate terms of address

• This is very rude

• It may cause unfriendly situation or arguments
• It can be very embarrassing, but it depends on the culture

8. Using bad language

• This is impolite not just in class but everywhere

• Can cause anger, losing concentration, and so on
• Unrespect for the teacher

9. Wearing inappropriate clothing

• It may be distracting. This is a school

• It is a personal choice
• It is important to be comfortable

Axial coding
Further examination of these comments seemed to suggest that they fell
into two groups: those that were accepting of the behaviour in question,
and those that were unaccepting. These might be set out as follows:
158  Developing Language Teacher Autonomy through Action Research

Item Behaviour Unacceptable Acceptable

1 Coming late Irresponsible OK if not habitual
Distracting
2 Using a phone Disrespectful OK for an emergency
Disturbing
3 Talking while Rude
someone Disrespectful
else is talking
4 Not paying attention Disrespectful
Distracting
5 Sleeping Disrespectful
Inappropriate in a
classroom
6 Eating or drinking Sometimes necessary
because of time or
health issues
7 Using inappropriate Rude
terms of address Embarrassing
Confrontational
8 Using bad language Causes anger
Disrespectful
9 Wearing Distracting Personal choice
inappropriate Need for comfort
clothing

From this analysis, we can see that a number of the behaviours (Items
3, 4, 5, 7, 8) received entirely negative comments, suggesting that there
was strong opposition among these students to talking while someone
else is talking, not paying attention, sleeping, using inappropriate terms
of address, and using bad language. One (Item 6) received only positive
comments, suggesting that eating and drinking in the classroom is accept-
able under some circumstances. Reactions to other behaviours (Items 1,
2, 9) were more balanced, suggesting a degree of tolerance towards com-
ing late, using a phone and standards of dress, according to circumstances
or personal choice.
Selective coding
When it comes to selecting an overarching theme, the topic is clearly
about behaviour, the context is a university classroom, and the partici-
pants are university students, so this might perhaps be expressed as:
Student perceptions of acceptable behaviour in a university classroom
 6  Analysing the Data 
   159

Summary
The information provided in this chapter is, of necessity, limited and
selective. For those who would like to know more about data analysis and
interpretation, we recommend a specialist text such as Dörnyei (2007).
Alternatively, taking a data analysis course can be time and money well
spent.
Although it is a common practice to hand one’s data over to a statisti-
cian to analyse, it is very empowering to be able to do it oneself, since
nobody else really understands what we have done, how or why we have
done it, and what we are trying to achieve. In other words, nobody will
analyse our data as well as we could ourselves once we have empowered
ourselves with the knowledge and experience of how to do it.

Task  Deductive analysis

Collect some written or verbal learner reflections on a particular aspect of
your lesson.
Transcribe them on a word document.
Prepare a set of categories from relevant studies.
Read the transcribed data carefully and code it against the pre-set catego-
ries identified.
Identify quotes from the data that would best represent each theme.

Task  Inductive analysis

Collect some written or verbal learner reflections on a particular aspect of
your lesson.
Transcribe them on a word document.
Begin open coding and create as many categories as you can.
Identify major themes among these categories.
Select quotes from your data that would best represent each theme.
160  Developing Language Teacher Autonomy through Action Research

Task  Analysis of observation and interview data

Decide on a particular focus for observing some of your learners in the class-
room: either through a video or in another teacher’s lesson.
Create a checklist or a semi-structured protocol as an observation tool.
For after the observation, prepare some questions that need further elabo-
ration by the learners.
Reconsider each observation focus with what they said in the post-­
observation interview.
Analyse the interview responses inductively.

Task  Analysis of quantitative data

Decide whether the data are numerical, ordinal, or nominal.
Calculate whether distribution is normal or not.
Calculate means or medians as appropriate.
Calculate reliability.
Calculate correlations and/or differences as appropriate.
Conduct any other appropriate tests.
 7
Discussing the Data

Interpretation
Action research is a professional development tool for teachers or practi-
tioners to investigate their classroom practices, not because they are defi-
cient, but because there is always room for further understanding and
development spaces. Action research requires systematic data collection
and analysis, which is then to be interpreted for deeper understanding
and reflecting on the issues under investigation in order to gain critical
insights. Interpretation requires asking so what questions, which helps
consider what the results might mean for the researcher and for the con-
text in which the study was carried out. Asking why questions will also
deepen the interpretation in a way to inform current and future teaching
practices.
However, there are no particular recipes for how to interpret the data.
It is a stage where researchers need to make meaningful pedagogical con-
nections among the emerging themes or results. Interpretation involves
an inductive process by which findings in the form of categories and pat-
terns are synthesized through meaning-making and developing insight-
ful, personal, but evidence-based explanations for what is practised in the

© The Author(s) 2017 161

K. Dikilitaş, C. Griffiths, Developing Language Teacher Autonomy
through Action Research, DOI 10.1007/978-3-319-50739-2_7
162  Developing Language Teacher Autonomy through Action Research

classroom. The degree of quality of the interpretations depends on the

researcher’s background, perspective, knowledge, and theoretical orienta-
tion as well as the intellectual skills brought to the task of interpreting
the data (Ary et al. 2010). Quality is ensured through ongoing reflection
on the outcomes of the data analysis to arrive at informed guesses and
data-driven inferences that might lead to gaining deeper insights into the
overall meaning of the research that has been undertaken. This is where
new meanings can be created out of the evidence which is presented
in the findings and analysis sections. Innovative meaning-making can
be achieved by developing and benefitting from the following positive
effects of action research:
–– adopting a research perspective concerning teaching practices
–– becoming free from academic domination and ownership of educa-
tional knowledge
–– generating knowledge that meets the needs of students, schools,
and societies by theorizing about practices through reflection
–– questioning practices and experimenting with new ideas that might
function effectively
–– improving teaching and learning in context by revisiting beliefs and
practices
These positive impacts can help build capacity to interpret findings in
order to induce pedagogical implications that inform practices. The fol-
lowing questions need to be asked:

• What meanings can I draw out from the findings?

• How can I connect the findings to form a coherent interpretation?

In interpretation, the following factors can be considered, which might

help with constructing new meanings:

Contextual Constraints and Conditions

Context plays a major role in understanding why results happen to be the

way they are. Contexts might shape the answers to the research questions.
 7  Discussing the Data 
   163

For example, let’s imagine that an action research identified low levels of
motivation among students particularly for participation in group activi-
ties. However, the students did not report much on why they had little
motivation to do that. Then, as a researcher, we need to try to understand
what the contextual information might tell us. We can also think about
the cultural aspects of the community. Since group work requires col-
laboration, respect, understanding, work share, and so on, it may some-
times be hard to establish group work. The students might be coming
from a culture in which such interaction patterns are not encouraged or
taught. This could guide us to understand the problem not from students
but from the critical links we make to the contexts. Such information
will help us particularly when we want to increase students’ motivation
to participate in group work. We might need to start by teaching basic
principles of group work rather than using group work as an interaction
pattern. As students learn how to behave in group work, we can see if
their motivation to engage in such activity increases.

Profile of Participants

Participants’ profiles might also provide several clues about how to inter-
pret the research findings. The age, gender, nationality, socio-economic
background, personal preferences, and so on, can help with the meaning-­
making process.

Data Collection Tools

The tools used to collect data could also play a role in the findings. For
example, some students may not be willing to write detailed responses
to open-ended questions or may not have time to do that, so their com-
ments and responses will not be enough to make new meanings out of
what they write. Such responses could make the research interpretation
superficial. This challenge could be overcome by using other ways to col-
lect data which involve more student participation as part of the regular
classroom activities.
164  Developing Language Teacher Autonomy through Action Research

Strategies for Reporting Interpretation

The interpretations of the data can be presented in various ways.
Developing different strategies may make interpretations understood
with more clarity as well as depth and breadth.

Concept Maps

Concept maps can show the relationships between emerging themes and
the issues investigated. These maps could show what interpretative per-
spective is developed from the findings. Linking the major issues drawn
from research can also display a professional and intellectual stance
towards the research results.

Bullet Points

Bullet points can be a clear way of communicating interpretations with a

focus on the highlights.

Visuals

Authentic pictures and photos can be used to support interpretations

which could create richness in meaning-making. These tools could pro-
vide insightful information on whether the study led to possible action,
and how the action and the interpretation can be connected.

L imitations and Directions for Ongoing

Research
In research, limitations refer to factors which must be considered when
assessing the meaning of the study. These may be methodological (such
as data collection tools or analysis procedures) or contextual (such as
the nature or number of participants, shortage of time, location, or the
­environmental conditions under which data was collected). It is possible,
 7  Discussing the Data 
   165

for instance, that the dual role as researcher and as teacher may lead stu-
dents or others to saying particular things that they think would be sat-
isfying. Or there might have been shortage of time when collecting the
data, which could influence the quality of the results.
A major limitation is that interpretations of action research may not
be generalized to the whole population from which the sampling was
formed. Since action research is commonly carried out with relatively
small numbers of participants, it is not reasonable to conclude that any
findings are necessarily applicable to a larger population. The extent of
the generalization is therefore confined to a particular sampling group.
Researchers also need to mention limitations which might involve
what has not been chosen, the other people outside the sample, and
methodologies used purposefully and others avoided because they were
not practical in the particular context. More specifically, it is necessary
to be aware of the procedures that are beyond the scope of a particular
research study. The discussion of the limitations will help the reader con-
textualize the research and understand it more appropriately.
But limitations should not be viewed only as negatives. In fact, any
study has limitations of one kind or another: it is simply not possible
to research every possible question everywhere perfectly and to univer-
sal satisfaction. The limitations can be turned into recommendations for
ongoing research which can be a valuable source of inspiration for other
researchers looking for a good topic.

Implications for Teaching/Learning
Action research is a way of reflecting critically upon teaching experiences
in order to improve practices. Action research also helps to develop prac-
tical and relevant implications which can be directly accessed as a set
of first-hand knowledge locally generated by action researchers. Among
the benefits of such research engagement are developing professionalism,
raising awareness, empowering the self, improving practice, encouraging
a reflective stance, engaging in collaboration, innovating and experiment-
ing with new practices, and developing autonomy. Developing through
these positive ­benefits stimulates professional growth adoption of respon-
sibility for what goes on in the classroom, which could directly impact
166  Developing Language Teacher Autonomy through Action Research

the teaching and learning process. Rather than teaching with the results
of professional researchers’ recipes or top-down curriculum decisions, or
coursebooks, teachers can enjoy taking the control of their own learning
and teaching process. Materials can be adapted and used in a way that is
more suited to the expectations of the students.
Practical or pedagogical implications relate to how the findings from the
study are operationalized in instructional practices. Since action research
involves systematic data collection and analysis followed by insightful
and evidence-based interpretations, it is useful to draw practical implica-
tions that could potentially work in the classroom. Implications could
be directed at different parties in the local context such as the self, teach-
ers, students, and staff depending on the focus of the research. Action
research can help understand what findings imply for others. In this way,
others can also benefit from the discoveries, and autonomy is developed
and strengthened.

Criticality
Starting from interpretation, reflection, and implications, it might be
necessary to have or develop abilities to create critical insights, perspec-
tives, and thoughts, which may not have been realized or generated so
far. Action research provides ample opportunities for the development of
a critical perspective. Banegas and de Castro (2016) refer to criticality as
“the practice of socially situated reflection and evaluation” (p. 1) which
involves creating multiplicity in perspectives while elaborating on issues,
including self-criticism. Criticality requires not only being negative about
issues but also questioning or challenging the existing assumptions and
developing a rational look into the questions in mind. One of the skills
that this book has aimed to develop is such a perspective from which ped-
agogical issues can be seen in ways that have not been considered before.
Deepening insights could open more unexplored and untouched ideas.
The untouched areas can only be accessed by systematic and insightful
reflection and research in combination, which is a potential that can be
achieved by engagement in action research. The critical look needed dur-
ing research plays a critical role in professional development. Research
 7  Discussing the Data 
   167

and interpretations of findings will urge elaboration of findings and

interpretations through constructive self-criticism. The critical perspec-
tive adopted may lead to improvement or modifications in practices. In
addition, strengths and weaknesses may be questioned, points to improve
may be identified, and awareness of changing beliefs may be developed
by means of the criticality in views. The evidence obtained from research
consolidates a critical stance, and a critical stance enriches learning and
development through engagement in research.

Questions to Consider
1. How can you ensure that your interpretations are evidence-based?
2. What are other ways of reflecting on your research?
3. Do you think you have to eliminate the limitations in your research?
4. How do you think teachers can be helped to sustain their engagement
in research?
5. What could be the ways of relating research and teaching/learning?
6. What strategies could you think of to become more critical in your
research?
7. How do you think action research can help you to become more
autonomous?

Example: An Action Research Study on Classroom Culture: Discussion

It is interesting that all of the items in the scale were rated on the medium
to very impolite side of the scale, with using inappropriate terms of address
and using bad language being considered most impolite (median  =  5).
Using a phone and talking while someone else is talking are also consid-
ered impolite (median = 4). Some participants felt that, given Istanbul’s
notorious traffic, coming late was sometimes unavoidable, while eating or
drinking in class might be all right under some circumstances as long as
it did not inconvenience others, since there are days when students start
early and finish late with no breaks between classes. A reasonable degree of
tolerance seemed to be extended to standards of dress, as long as it was not
“embarrassing” or “distracting”, and the question was also raised by some
students that it was not easy to achieve a definition of what was “appropri-
ate”, a point that was also made regarding terms of address (these points
168  Developing Language Teacher Autonomy through Action Research

were discussed in class as part of a follow-up to the study). Several other

impolite behaviours were suggested, including failure to bring the correct
material, chewing gum, and attending to personal grooming during class.
Another interesting feature of the results is the unanimity across
learner variables. We might have expected males and females to have dif-
ferent opinions about some of these behaviours, but this did not prove to
be the case. Furthermore, although there was one statistically significant
difference according to nationality, suggesting that Turkish students are
less concerned about punctuality than their international classmates, the
remainder of the differences according to nationality were not significant.
Implications
Once the results had been found and the study written up, it was distrib-
uted to the class for discussion. There was some amusement with some
of the findings (e.g. about eating, drinking, or sleeping in class), and
some embarrassment from some who recognized impolite behaviours in
themselves (e.g. consistently coming late and entering in such a way as to
distract classmates, using phones during class). In the weeks following the
study, there was some evidence that students made some effort to control
their own negative behaviour (e.g. by being more punctual, turning off
phones when they arrived in the room) and some also seemed to be more
inclined to attempt to control classmates for impolite behaviour (e.g. if
they were talking while someone else—either the teacher or one of their
own classmates during a presentation). In other words, this study seemed
to have some effect with raising students’ own awareness of the effect of
their behaviour on others, and to result in more willingness to engage
in polite behaviour, which, in turn, seemed to have a positive effect on
classroom dynamics.
Suggestions for further research:
There are a number of ways in which this study might be extended:

• An interesting direction would be to give the survey to teachers and

compare their perceptions with those of students.
• Also, interviews could be conducted for a more in-depth qualitative
perspective.
• Some of the other behaviours the students considered impolite might
be added to the questionnaire
 7  Discussing the Data 
   169

Conclusion:
Overall, there is agreement that four out of nine (44.4%) of the behav-
iours listed in the questionnaire are impolite or very impolite. Students
did not express strong opinions about the remaining five items (55.6%),
but none of the items was rated polite. Perhaps if students are made aware
of these findings it might help to promote a more harmonious classroom
atmosphere which is likely to promote more effective learning.

Task  Write a discussion section for your study in which you

1. Interpret your findings
2. Identify any limitations
3. Suggest useful directions for ongoing research
4. Suggest implications for the teaching/learning process
 8
Example Studies

Although researchers often want to achieve something “original”, in fact,

replication of an existing study, whereby the former study is repeated in
another context in order to test its generalizability beyond the original
setting, is a valid form of scientific enquiry (Porte 2012). From a novice
researcher’s point of view, having an already established structure to fol-
low can provide extremely helpful scaffolding, while also providing prac-
tice and building confidence and “savoir faire”.
With this in mind, this volume will provide several potentially repli-
cable examples of action research of different types (quasi-experimental,
a case study, a questionnaire study, and a mixed-methods study) in order
that others who might be interested might take the ideas and apply them
in their own situations.
Replicating a published study in different contexts offers two benefits.
First, the researcher may not have to design the study, which is good
especially for novice researchers. Replication experience might also help
teachers gain confidence in conducting research, which allows them to
learn the process of setting hypotheses or asking questions as well as col-
lecting and analysing data. They can also learn more about how to display
the findings in tables, figures, or charts as appropriate. Second, ­replication

© The Author(s) 2017 171

K. Dikilitaş, C. Griffiths, Developing Language Teacher Autonomy
through Action Research, DOI 10.1007/978-3-319-50739-2_8
172  Developing Language Teacher Autonomy through Action Research

studies increase the value of research (Creswell 2014) because if the results
are supported or contradicted in different contexts, this could add to or
reduce the generalizability of the conclusions. The knowledge gained in
different research studies can therefore add to the depth and breadth of
accumulated knowledge.

A Quasi-Experimental Study
 he Place of Extensive Reading in the Language
T
Classroom

Arife Koç

Abstract
This quasi-experimental study took place in an existing classroom using
an intervention to investigate whether extensive reading (ER) made a dif-
ference to students’ results, which were examined by means of a pre-test
and a post-test. The results indicated that the class which received the
intervention was significantly more successful than the control group.

Introduction

According to the input hypothesis (Krashen 1982), reading is very

often considered an effective source of comprehensible and meaningful
input which is a sufficient condition for the unconscious acquisition
of the target language (TL). The reading hypothesis (Krashen 1993)
provides further support for reading as a crucial means of language
instruction by stating that the more we read in a TL the greater our
vocabulary will be.
As Nation (2009) puts it, reading provides meaning-focused input
where learners focus on the meaning of the text and there are only a
few language features beyond their proficiency level. Extensive reading
(ER) is an important source of meaning-focused input which means
engaging students in large amounts of reading. Reading also provides
 8  Example Studies 
   173

the learners with a context for language-focused learning which involves

focusing on the language items deliberately. Furthermore, reading is a
source of fluency development which takes place when there are almost
no new language items to master and when the goal is to make the
best use of what has been mastered before. Fluency development in
reading, as well as in writing, is often ignored because teaching and
learning something new usually has the priority in language courses.
However, fluency in reading is essential in developing reading compre-
hension skills in the TL since increased fluency assists comprehension
and ultimately improves language learning attitude. When only 2–5%
of the vocabulary or the grammar items are unknown, this provides
meaning-focused input conditions. On the other hand, having almost
no unknown vocabulary or grammar structures provides fluency devel-
opment conditions (Nation 2009).
Since it tends to be relaxed, and students are allowed to pick materials
based on their language proficiency level and interests, ER is sometimes
called pleasure reading (PR, Day and Bamford 1997), free reading (FR,
Krashen 2004), sustained silent reading (SSR, Kelley and Clausen-Grace
2006), or drop everything and read (DEAR, Nation 2009). ER has been
drawing growing attention from many English as a second language
(ESL)/English as a foreign language (ELF) educators and researchers for
being an effective and efficient means of instruction in English language
teaching (ELT) (Grabe 2010; Nation 2009).
Learning via ER is mostly incidental, that is, learners focus on the
story rather than the language items. That is why quantity of input which
involves adequate opportunities for substantial vocabulary repetition is
required to foster learning gains. The quantity of input is meant to be
around 500,000 running words in a year, which is equal to about 25
graded readers per year (Nation 2009). ER could take place within or
outside the class time, which in both cases require careful planning as
well as attentive monitoring. In other words, ER is defined as a procedure
of providing students with “the time, encouragement, and materials to
read pleasurably, at their own level, as many books as they can, without
the pressures of testing or marks” (Davis 1995, p. 329).
174  Developing Language Teacher Autonomy through Action Research

Literature Review

There have been numerous researchers who have studied the effects of
ER on language learning in ESL/EFL contexts, and many studies have
found that ER benefits the language learners in numerous ways includ-
ing their reading proficiency, reading habits, reading fluency, vocabulary
retention, as well as writing and spelling (Nation 1997). Besides the gains
in different language skills, learners often experience delight and positive
feelings in language learning as ER encourages them for further reading
and study (Krashen 2004).
In the 1980s, Elley and Mangubhai (1983) carried out a compre-
hensive study. They investigated the impacts of an eight-month read-
ing programme (Book Flood) on second language receptive skills. The
participants were 380 fourth and fifth (9–11 years old) graders from 12
schools in Fiji. The study compared the two Book Flood groups, that is,
the SSR Group and the Shared Book Experience (SBE) Group, and the
control group. With the SBE group, the students were read aloud five
high-interest stories by the teacher and were asked to complete follow-up
activities such as art work or roleplaying. The students in the SSR group
read self-selected readers silently for about 30 minutes in class with no
follow-up activities. The students in the control group continued their
regular audio-lingual language programme. To assess reading comprehen-
sion improvement, the researchers applied a Reading Comprehension test
with 32 multiple choice items as the pre-test and post-tests. The results
indicated that the experiment/“Book Flood” groups achieved more than
twice the score achieved by the students in the control group in reading
comprehension skills, and they also showed a 15-month reading growth
during this 8-month reading-based programme.
In her study, Tum (1995) was mainly interested in the effects of ER
on vocabulary improvement. She also analysed data from achievement
tests to figure out the impact of ER on reading comprehension. The
participants were 100 EFL students from YADIM (Foreign Language
Center at Çukurova Unıversity, Turkey). They were divided into four
groups: two experimental and two control groups. At the end of the
eight-week ER outside the classroom, the findings of the post-test
results showed that there was a significant difference when the averages
 8  Example Studies 
   175

of the experimental and the control groups were compared in reading

comprehension. The average of all four groups was 43 on the pre-test.
The control groups increased their average to 53.5% while the average
was 72% for the experimental groups suggesting that ER significantly
influenced their reading comprehension.
Yang (2001) carried out a study to examine the effects of reading nov-
els, especially mystery novels, on English proficiency level and motiva-
tion. The participants were 120 adult learners from Hong Kong. Besides
their regular textbook, the experimental group read two mystery novels.
They had discussions on the plots, characters, and the social issues of
the novels. They also wrote their opinions about the social issues they
found while reading the novels every other week. The control group
only covered the regular textbook. The study lasted for 15 weeks with
three-hour weekly sessions. As pre- and post-tests to assess the language
improvement, 100-item sentence completion tests with multiple-choice
items were administered. When the results of the pre- and post-tests
were compared, it was concluded that the experimental group students
showed a greater language gain than the students in the control group.
The improvement in language proficiency for the experimental group was
almost twice as much as the control group.
In his comprehensive research, Yanmaz (2015) analysed the effects of
ER on reading comprehension, vocabulary and grammar knowledge,
and attitudes towards reading with forty ninth grade students. The study
lasted for 10 weeks, during which students were to read one book per
week followed by class discussions when the reading was completed. The
books were assigned by the teacher. The pre- and post-test results sug-
gested that the ER treatment significantly improved the reading compre-
hension skills of the participants in the experimental group.

Research Question

The current study was designed to find the answer to the following
research question regarding ER and the relationship to foreign language
achievement:
176  Developing Language Teacher Autonomy through Action Research

What is the effect of ER on the overall language achievement in the for-

eign language classroom?

Research Context

Setting

The study was conducted at the Preparatory School of a private university

in Istanbul, Turkey.

Participants

As the participants of the study, there were 37 Turkish EFL students in

two classes, 17 males and 20 females. They were all Turkish native speak-
ers, aged 18–19 years old. They had been learning English intensively
for about six months before the study and were at B1 level according to
the Common European Framework (CEF). It was their fourth week in
the B1 module when they took the Achievement Exam 2 as a pre-test to
identify the 2 classes out of 24 with the closest results and taught by the
same teachers. Class 1 was assigned to be the control group (n = 19), and
Class 2 (n = 18) was assigned as the experimental group.

Data Collection

Materials

A total of 81 graded readers of various genres and grade levels were pro-
vided for the in-class ER programme with the experimental group. The
readers were from different publishers’ series of readers, such as Oxford
Bookworms Library, Penguin Readers, Heinemann Guided Readers, and
Macmillan Readers.
 8  Example Studies 
   177

Instruments for Collecting Data

Achievement Exam 2, consisting of listening, reading, grammar, vocabu-

lary, and writing sections, was administered to all the students at the
school, and this was used as the pre-test to see the English achievement
level of both the control and the experimental groups. Achievement
Exam 5, also consisting of listening, reading, grammar, vocabulary, and
writing sections, was considered to be the post-test to identify if there was
any impact of the treatment. No pre-reading tasks, follow-up activities,
or book reports were asked from students in order to make sure that the
treatment was stress-free.

Procedure

Before the study was conducted, necessary approval was taken from the
administration, because the intervention would be conducted within the
class time, which meant having to catch up with the syllabus whose con-
tent the students were to cover for their achievement tests every other
week.
The procedure used for the ER included the following:

• ER should be undertaken for at least 30 minutes every day in the

scheduled class time.
• Students should only read their readers, not any other work.
• When students finish a book, they should note the name and level of
the book next to their names in the log.
• Students themselves should pick another book of their interest and
level when they finish a book.
• Students should not take the books with them after the session.
• At the end of the sessions students should put the books back in the
box.
• The box and the log should be placed in the locker after the session.

The graded readers box was kept locked in a cupboard very close to
the experimental classroom, and the instructor would take the box to the
178  Developing Language Teacher Autonomy through Action Research

class when the ER sessions were scheduled and put it back in the same
locker when the session was over. Students did not keep the books with
them; they were given the books they had been reading at the beginning
of the session and were asked to put the books back in the box at the end
of the session. They would pick another one of their interest and level
when they were finished with a book and asked to note down the name
and level of the book when returning it on a log which was kept to track
the number and level of the books read.
Students were not allowed to do any other work during the sessions.
The teacher also read to set a good example when the students were read-
ing. Students were told to pick the readers below the level they assumed
they were at first. They were advised to read the very first paragraph(s)
of the book to see how comfortable they felt in understanding the lan-
guage, remembering that there should be no more than 2–5 unknown
words in about 100 words. They were also told to read several books of
the same level before they felt competent to go on reading higher-level
graded readers.
The treatment took seven weeks with a couple of days off due to
national holidays. At the end of the treatment, a total of 118 books were
read with an average of 6.2 books per student. Although there were some
slow readers who covered as few as three or four books in seven weeks,
there were also several students who could read as many as nine books in
the same period.
At the end of the seventh week of the ER programme, Achievement
Exam 5 was scheduled to be given. This test was taken as the post-test of
our treatment to investigate whether there was any difference between
the test results of the ER group and the control group.

Data Analysis

The results obtained from the pre-test (Achievement Exam 2) and post-­
test (Achievement Exam 5) for both the control group and the interven-
tion group were analysed using the statistical software program SPSSv.20.
The results of the achievement exams were analysed for normality of
distribution and means. Independent samples t-test analyses were done
   
8  Example Studies  179

to identify whether or not the ER intervention within the classroom

resulted in any significant difference in students’ overall English achieve-
ment level.

Results

To identify which tests to apply, the normality of the distribution of the pre-
test total scores was checked. The pre-test scores were found to be normally
distributed. Given normal distribution, pre-test means were calculated and
an independent sample t-test was applied to the control and the experi-
ment groups’ pre-test results in order to identify whether the difference
between the means of the two groups was significant or not. As can be seen
in Table 8.1, the pre-test means of the control group (M = 68.26) and the
experimental group (M = 68.78) were very close to each other. The analysis
of the results with the independent sample t-test (p = 0.90) also showed
that there was no statistically significant difference between the groups in
terms of overall English achievement level before the ER treatment.
After the seven-week ER programme conducted within the class time
for about 30 minutes per school day, students were given Achievement
Exam 5, which was regarded as the post-test in this study. The post-test
scores were found to be normally distributed. Since the distribution was
normal, the means of the pre-test and the post-test were compared in
order to investigate if the ER treatment had any impact on students’ lan-
guage achievement (see Table 8.2).
The post-test mean of the experimental group (M = 74.44) was
numerically higher than the mean (M = 65.89) of the control group (see
Table  4). To test if the difference was statistically significant, an inde-
pendent samples t-test was performed. This indicated that the difference
between post-test results of the treatment group and the contrast group
was statically significant (p = 0.03).

Table 8.1  Group statistics of pre-test total scores

Group N Mean Std. deviation Std. error mean
Pre-total Control group 19 68.26 11.82 2.71
Experiment group 18 68.78 14.17 3.34
180  Developing Language Teacher Autonomy through Action Research

Table 8.2  Group statistics of post-test total scores

Group N Mean Std. deviation Std. error mean
Post-total Control group 19 65.89 11.99 2.75
Experiment group 18 74.44 10.27 2.42

Discussion

According to the results of this study, ER with pressure-free features

appears to have positive effects on language achievement levels. As we
can see from the results, the ER group increased the achievement test
means from M = 68.78 to M = 74.44, while the control group did not
show any increase (indeed, there was a slight drop in the mean from M
= 68.26 to M = 65.89). The difference between the pre-test results of the
groups was not significant prior to the treatment, whereas the post-test
comparisons of the results of the groups indicate that the treatment of
ER seemed to have a considerable effect on the students’ overall language
learning achievement.
It should be noted that there were some concerns about the struc-
ture of this research at the beginning. Because the ER would take place
within the classroom during instruction time, students in the experimen-
tal group would receive less school instruction than the control group,
and the content of the instruction was to be asked in the achievement
exams; therefore there had to be some reduction in time devoted to teach-
ing the subject material of the course. These concerns remained until
Achievement 5 (post-test) results were announced, when it was dem-
onstrated that the concerns were unnecessary, since assumptions to the
contrary happened to be true: students in the ER group did much better
than the group who were bound to a scheduled curriculum. Whatever
the students gained from ER seems to have outweighed what they missed
from their syllabus. Maybe, it could be suggested that more exposure
to the language, even though it was simplified, fostered what has been
learned and reinforced the basic structure of the language and formed a
stronger basis for the learners to take the next step of language learning
more confidently.
 8  Example Studies 
   181

Implications

In this study, although there were some concerns at the beginning, an

ER programme was deployed in the classroom, which meant devotion of
some class time. However, there was no need to worry, as results showed
that reading compensated for what students missed and added more to
their competence. So, inserting the ER into the classroom should not be
concerning, as the current study suggests.
As Krashen’s Affective Filter Hypothesis suggests, during the ER ses-
sions, students were experiencing a language exposure with less stress
since they were not asked to do any tests, quizzes, or fill in any review
or report papers, which was one of the distinctive features of this study.
Besides, they were given the chance to pick the text they liked the content
of and felt more comfortable with the level, which again matches with
Krashen’s Comprehensible Input Hypothesis. Thus, the present approach
to ER suggests that we should just provide our students with an environ-
ment for them merely to read. This actually puts almost no burden on
either the students or the teacher, although the gains are impressive.
In order to have success in an ER programme, there are certain require-
ments to fulfil. First of all, there needs to be a large amount of reading—
one book per week is recommended—of materials which are lower than
the learners’ actual level of language. The reading materials should be of
a variety of topics and genres. It should be the students who pick what
to read, although teachers can provide guidance if asked by the learn-
ers. Reading should be fast, rather than slow, so any task which slows
students’ reading pace should be avoided. Reading should be individual
and silent. The teacher, in order to be a role model and a good example
of a reader, should also read with the students, preferably a book from
the class set.
However, when applying ER, in order to be able to keep the interest
of the students high, there should always be a good variety of resources.
Towards the end of the study, there were some demands for more books,
which meant students would lose interest in reading if the study contin-
ued for some more time with the same set of readers. So it would be a
good idea to add new reading materials to the pack to keep up the inter-
est and enthusiasm.
182  Developing Language Teacher Autonomy through Action Research

Moreover, some participants may tend to take the ER sessions as free

time activity since reading is mostly considered so, and may be reluctant
to read but prefer to do some other assignments. Therefore, participants
must be persuaded to take these sessions seriously because this is not a free
time activity but a course requirement. So, there may need to be a good
monitoring for the sake of the application. Besides, monitors should also
read themselves in order to model the desirable behaviour.
Despite the consistently positive results of ER programmes, there still
seem to be doubts as to its effectiveness: study after study says it works,
but very few language programmes have adopted it (Krashen 2007).
Thus, with another study which approved ER, this low-burden but high-­
gain approach to reading should be implemented in foreign language
classrooms. Reading can be a good source of not only learning but also
enjoyment. It provides a source for both establishing previously learned
grammar and vocabulary and learning new language items. Moreover,
with the feeling of achievement and success in using the language, it can
yield further enthusiasm in learning more. As the learner improves in see-
ing reading as a means of gaining world knowledge, the enjoyment can
increase, which accelerates further learning.

Limitations and Suggestions for Further Research

The number of subjects in the control group and the treatment group was
only 37. For more reliable research findings the number of participants
needs to be increased. The period of ER treatment was also very short.
More significant findings could be reached if there could be more time
when implementing the ER programme.
For ER, there should always be a good variety of resources. Towards
the end of the study, there were some demands for more books, which
meant students would lose interest in reading if the study continued for
some more time with the same set of readers. Besides readers, since the
aim is to make the students read what they like to read, there should
have been more variety in terms of not only genres but also content. For
instance, it would have been more appealing if there were not only read-
ers but also some magazines, comics, graphic novels, and so on, about
 8  Example Studies 
   183

different topics like music, sports, cinema (stars), and so on, in the read-
ers’ box in this study.
The ER study had a quasi-experimental research design. Since the
experiment was to be conducted in an institution during the school day,
participants could not be taken out of their original classes, so groups
were assigned as classes rather than with random sampling. More reliable
and valid results could be yielded with randomly assigned groups.
Since our learners are so much involved in technological improve-
ments nowadays, ER integrated with technology or Internet could be
more appealing to language learners. So, the effects of e-books or any
other applications regarding reading should also be tested.

Conclusion

This study has clearly shown that ER can be and should be implemented
in a language classroom. Anyone concerned with language education,
teachers, administrators, or students should open-mindedly consider
what benefits ER would bring to the language learning experience of the
students. ER should be embedded in the classroom teaching, as learn-
ers would benefit from it so much that it should not be left outside the
classroom. ER should not be taken only as a free time activity, because it
really is a serious, useful learning experience which should be valued and
introduced as an integral aspect of the classroom environment.

References

Davis, C. (1995). Extensive reading: An expensive extravagance? ELT

Journal, 49, 329–335.
Day, R., & Bamford, J. (1997, May 21). Extensive reading: What is it?
Why bother. Retrieved from http://jalt-publications.org/old_tlt/files/97/
may/extensive.html
Elley, W., & Mangubhai, F. (1983). The impact of reading on second
language learning. Reading Research Quartely, 19(1), 53–67.
184  Developing Language Teacher Autonomy through Action Research

Grabe, W. (2010). Fluency in reading – Thirty-five years later. Reading

in a Foreign Language, 22(1), 71–83.
Kelley, M., & Clausen-Garce, N. (2006). R5: The sustained silent
reading makeover that transformed readers. The Reading Teacher, 60,
148–156.
Krashen, S. (1982). Principles and practice in second language acquisi-
tion. Oxford: Pergamon Press.
Krashen, S. (1993). The power of reading: Insights from the research.
Englewood: Libraries Unlimited.
Krashen, S.  D. (2004, April). Free voluntary reading: New research,
applications, and controversies. Paper presented at the RELC conference,
Singapore.
Krashen, S. (2007). Extensive reading in English as foreign language
by adolescents and young adults: A meta-analysis. International Journal of
Foreign Language Teaching, 3(2), 23–29.
Nation, I.  S. P. (1997). The language learning benefits of extensive
reading. The Language Teacher, 21, 13–16.
Nation, I.  S. P. (2009). Teaching ESL/EFL reading and writing
(pp. 49–60). New York: Routledge.
Tum, G. (1995). A study on vocabulary improvement through extensive
reading. Unpublished M.A. dissertation, University of Cukurova, Adana.
Yang, A. (2001). Reading and the non-academic learner: A mystery
solved. System, 29, 451–466.
Yanmaz, S. (2015). The effects of extensive reading on enhancing reading
comprehension, vocabulary recognition, grammar competence and attitudes
of the A1 level learners of English. Unpublished M.A. dissertation, Cyprus
International University, Nicosia.

Why ESP? A Case Study of a Business English Class

Carol Griffiths

Abstract
Over the years there has been a great deal of controversy regarding the
role of English for Specific Purposes (ESP) in English Language Teaching
 8  Example Studies 
   185

(ELT). Proponents point to the potential for ESP to provide motivation,

since students can align the courses with their own goals. Opponents,
however, insist that ESP courses are no different from General English
(GE) courses except for the superficial distinction of the topics on which
they are based. Amid this debate, often conducted at an academic level
somewhat removed from the realities of actual classrooms, the question
of what students see as their needs is frequently sidelined. This study
takes a grass-roots look at the case of an ESP course as exemplified by a
Business English course offered at a private tertiary institution. Student
motivation is investigated, firstly along traditional lines (intrinsic, extrin-
sic, integrative, instrumental), though it is also pointed out that these
distinctions may be overly simplistic. Students are then asked to identify
their needs, which, in spite of the “specific” nature of the course, tend to
be expressed very much in linguistic terms. At the end of the course, stu-
dents’ levels of satisfaction are rated and positive and negative reactions
to the course are evaluated. These findings are discussed with implica-
tions for the teaching/learning situation. Finally, suggestions are made
for ongoing research.

Introduction

Over the years, ESP has been defined in various ways. The difference
between GE and ESP is explained by Widdowson (1983, pp. 108–109)
in these terms: “In ESP we are dealing with students for whom the
learning of English is auxiliary to some other primary professional or
academic purpose. It is clearly a means for achieving something else
and is not an end in itself.” Hutchinson and Waters (1987, p. 54) agree
that it is the goal, or, as they call it, the “target situation [which] distin-
guishes the ESP learner from the learner of General English”. Students
study ESP not primarily because they are interested in English, but
because they need English for some particular objective. They may want
to go to university, they may want to improve their job prospects, or
they may want English to facilitate business negotiations. However, as
Brinton et al. (1989) suggest, merely contextualizing lessons organized
around vocabulary, grammar, or functions is not enough; it is for this
186  Developing Language Teacher Autonomy through Action Research

reason that “proponents of English for Specific Purposes (ESP) note

that for successful language learning to occur, the language syllabus
must take into account the eventual uses the learner will make of the
target language” (p. 3). Johns and Dudley-Evans (1991, p. 305), how-
ever, believe that ESP has requirements that are “specialized or unique”.
These requirements, they maintain, set ESP apart from other English
language courses.
According to these definitions, then, ESP is different from GE
mainly in its focus. Instead of being focused on developing language,
an ESP course focuses on the purpose for which the language is being
developed. This focus gives a sense of immediate usefulness and pro-
vides a motivation for learning language, even though language is not
the ostensible primary purpose of the course. As a result, a vast range
of courses have sprung up to meet (some would say cash in on) this
demand. There is English for Commerce, English for Science, English
for Technology, English for Medicine, Business English, English for
Secretaries, Communications English, English for Tourism, English
for Librarians, English for Tradespeople—the list could go on and on.
Some of the more unlikely sounding variations mentioned by Cervi
(1993, p.  14) include “English for Miners” and “English for Scuba
Diving”.

Literature Review: Characteristics of ESP

Advocates of ESP point to several aspects which they believe are special
to ESP and which set it apart from GE courses. Strevens (1988, p. 1), for
instance, includes among the “absolute” characteristics of ESP courses
that they are
–– designed to meet specified needs of the learner
–– related in content (i.e. in its themes and topics) to particular disci-
plines, occupations, and activities
–– centred on the language appropriate to those activities in syntax,
lexis, discourse, semantics, and so on, and analysis of this discourse
–– in contrast with “General English”
 8  Example Studies 
   187

Other areas commonly mentioned as having special characteristics

with ESP courses include the following:

Needs Analysis

Johns and Dudley-Evans (1991, p.  299) tell us that “Throughout its
history, ESP practitioners have been pre-occupied with learner needs”.
Identifying learner needs, they believe, is an essential precursor to course
and materials design. Schmidt (1981, p. 199) agrees that when design-
ing ESP curricula “needs assessment is recognised as a desirable star-
ing point”. There  are two elements to needs analysis: Target Situation
Analysis and Present Situation Analysis. The Target Situation Analysis
includes discovering what students need to be able to do at the end of
the course, the language skills and functions and the degree of language
proficiency that will be expected, and how much subject knowledge will
be required. The Present Situation Analysis includes examining students’
present situations and current ability levels.

Materials

Since ESP courses are claimed to have a specific focus (be it science,
business, tourism, or whatever), it would seem to follow that the mate-
rials used for such courses need to relate to this focus. ESP students
commonly expect that the materials they are using relate to their tar-
get situation in some way. Another well-established trend in ESP is
that of using authentic materials. Robinson (1991, p.  54) notes: “A
key concept…and one felt to be particularly relevant for ESP, is that of
authenticity”.

Assessment

Hutchinson and Waters (1987, p. 144) make the point that assessment
is very important in ESP since “ESP is concerned with the ability to per-
form particular communicative tasks. The ability to assess proficiency is,
188  Developing Language Teacher Autonomy through Action Research

therefore, central to the whole concept of ESP”. Because ESP courses usu-
ally have more exactly specified objectives than is possible in GE courses,
assessment may be brought more sharply into focus in ESP courses than
in GE courses.

Language

The concept that language changes according to the context in which it is

used has given rise to a commonly held belief that specific contexts gen-
erate distinct registers or “formally differentiated varieties of a particular
language” (Widdowson 1974, p. 28). As a result of this belief, many ESP
courses have been based on the assumption that a particular kind of lan-
guage needs to be taught to prepare students for particular situations. As
Robinson (1991, p.  27) points out “for many people vocabulary, par-
ticularly specialist vocabulary (or terminology) is a key element of ESP”.
Students who opt for ESP courses often give anxiety over vocabulary as
one of their major reasons for doing so. The concern with how language
is used has led to functions having a central place in many ESP syllabi,
since certain language functions (such as explaining, describing, report-
ing) are sometimes thought to be more typical of some contexts than
they are of others. For example, an examination of a textbook such as
International Business English (Jones and Alexander 1989) reveals that it
covers functions such as “meeting people”, “requesting”, “complaining”,
“apologizing”, and so on.

Content

The content of ESP courses can be a thorny issue for the teacher. Few spe-
cialist English teachers are also scientists or business managers or experts
in tourism. Yet suddenly they must teach not just language, but language
embedded in a topic about which, quite possibly, their students know
much more than they do. A criticism sometimes levelled at ESP is that
 8  Example Studies 
   189

the process of learning is sometimes neglected as a result of the emphasis

on content. This approach emphasizes the means rather than the ends
and selects content to promote the requirements of the learning pro-
cess rather than attempting to adapt the process to the demands of the
content. In a process-oriented approach, according to Hutchinson and
Waters (1987, p. 70), the emphasis “is not on achieving a particular set
of goals, but on…helping learners to develop skills and strategies which
will continue to develop after the ESP course itself ”. It is important to
remember that an ESP course is basically an English course using special-
ist content as a vehicle for the process of learning language rather than
a content course which happens to be in English. Pinto da Silva (1993,
p. 40) reassures the language teacher that “it is now widely accepted that
the ESP teacher should not be expected to be an expert in the student’s
specialty”.

Face Validity

ESP courses are sometimes criticized because many of the characteris-

tics which are often proposed as distinguishing ESP courses from GE
courses (e.g. the vocabulary, the functions) are in fact no more unique
to ESP than they are to any other language teaching situation. And yet
ESP courses remain very popular. This is possibly because of a percep-
tion that an ESP course is more related to a student’s eventual goals than
is a GE course. This perception provides a motivation which may be
absent in a GE course whose content and aims may appear irrelevant in
terms of the student’s ultimate objectives. This perception is known as
face validity, and is based on the fact that “subject-specific materials look
relevant” (Hutchinson and Waters 1987, p. 166). Although it is possible
to argue that professional ethics dictate that a course needs to do more
than merely “look relevant”, Hutchinson and Waters (1987, p. 166) tell
us: “These factors should not be discounted. They are very important to
the learners.” If students believe that what they are studying relates to
their eventual goals, they are likely to be more interested and motivated
190  Developing Language Teacher Autonomy through Action Research

to succeed than they would be if the subject of their study seems irrel-
evant to their target situations.

Methodology

There is disagreement among ESP experts on the subject of methodol-

ogy. According to Johns and Dudley-Evans (1991, p. 305), “ESP requires
methodologies that are specialized or unique”; Hutchinson and Waters
(1987, p. 142), however, disagree. They express the opinion that “There
is nothing specific about ESP methodology. The principles which under-
lie good ESP methodology are the same as those that underlie sound
ELT methodology in general…The teacher who has come to ESP from
General English need not think that a whole new methodology must be
learnt.”

Research Questions

In order to further explore the controversies surrounding the question of

ESP, a study was set up to investigate the following questions:

1 . What motivates students to choose an ESP course?

2. How do ESP students view their needs?
3. Are students’ needs met on an ESP course?
4. What are the positive/negative aspects of an ESP course?

Of these questions, the first two were asked at the beginning of the
course, while the third and fourth questions were asked at the end.

Research Methodology

The study employed a case study approach. Case studies are not always
easy to define. A distinguishing feature is that they use specific cases to
generalize about the wider population. Cases may be individuals or par-
ticular groups, they may use participant or non-participant o­ bservation,
 8  Example Studies 
   191

and they may be conducted in natural or artificial settings. They are

usually longitudinal, often involve qualitative rather than quantitative
data and analysis, and may be structured or unstructured depending on
the level of control desired by the researcher. Case studies often derive
theory from data rather than the other way around, a process known as
“grounded theory”. It is assumed that the case being studied is typical of
a type, and that, therefore, it will be possible to generalize from the case
being studied to other similar cases. These generalizations are “based on
the assumption that the information gathered on a particular individual,
group or community will also be true of the other individuals, groups
or communities” (Richards et al. 1992, p. 47). A great strength of case
studies, according to Adelman et  al. (1976), is their reality. Although
the data gathered from case studies is often difficult to organize (partly
because there is often so much of it), they often hold a reader’s attention
and are easy for a reader to relate to. This is possibly because case studies
deal with real people in real situations, as distinct, perhaps, from experi-
mental groups which often seem rather removed from reality. Case stud-
ies, according to Adelman et al. (1976), are a step to action. The insights
gained from case studies can be used for teacher development, for feed-
back, for formative evaluation, and for policymaking, and, as such, they
are an especially useful method for an action researcher. Furthermore, as
Johnson (1992) puts it, case studies are “not only informative but also
involving and entertaining” (91).

Instrumentation

Research questions 1 and 2 (“What motivates students to choose an ESP

course?” and “How do ESP students view their needs?”) were asked at the
beginning of the course. This involved handing out a question sheet on
which the two questions (“Why did you choose this course” and “What
do you want to get out of this course?”) were written with space allowed
for student responses (see Appendix A).
For research questions 3 and 4 (“Are students’ needs met on an ESP
course?” and “What are the positive or negative aspects of an ESP course?”)
a reasonably simple three-item questionnaire was constructed. The first
192  Developing Language Teacher Autonomy through Action Research

question was rated on a five-point scale while the second was divided into
two sections (positive and negative) and space allowed for student com-
ments (see Appendix B). This was used at the end of the course.
A questionnaire was chosen as the basic instrument because “they are
quick and easy to administer…and are almost completely nonthreaten-
ing when administered using paper and pencil (or computer) under con-
ditions of confidentiality” (Oxford 1996, p. 33). Furthermore, since they
are less dependent on the researcher’s interpretation than may be the case
with other research methods (such as interviews), questionnaires provide
a “relatively objective platform for launching further probes” (Nunan
1992, p. 143).

Research Context

Setting

The study reported here took place at a private tertiary institution in

Auckland, New Zealand. Students came to this institution from a num-
ber of different nationalities to study for a range of courses, including
English language as well as a number of diplomas and degrees in Business,
Tourism and Information Technology. Quite often, the English language
students were studying in order to qualify for one of the diplomas or
degrees, and these students often reported finding the GE courses “bor-
ing”. It was for this reason that it was decided to set up a Business English
class to provide motivation for those wanting to go on to further study.
The class also operated as a pilot for a Business English course to be
operated separately from the GE course with its own NZQA (New Zealand
Qualifications Authority) accreditation. For this purpose, the documenta-
tion required (such as course materials, assessment procedures, etc.) was
collected during the course for submission to NZQA at the end.

Participants

There were 12 students who signed up for the new Business English class.
The characteristics of the participants were as follows:
 8  Example Studies 
   193

Nationality

Three different nationalities were involved in the class:

Three were Japanese

Four were Taiwanese
Five were Korean

Age

There were no teenagers (i.e. nobody under age 20).

Six were in their 20s

Four were in their 30s
Two were in their 40s

Gender

The class was evenly divided:

Six men
Six women

Data Collection

Permission to conduct the study was obtained from the institution’s

research ethics committee. Students were informed about the nature of
the study and written permission was obtained.
The first questionnaire  (see Appendix A) regarding motivation for
taking the Business English course and students’ perceived needs was
handed out during the first lesson following introductions and course
information. Students were given time to think about their responses and
write their ideas on the questionnaire form (about 15 minutes) before the
forms were collected for later analysis.
194  Developing Language Teacher Autonomy through Action Research

The second questionnaire (regarding how well the course met their
needs and positive and negative aspects of the course - see Appendix B)
was handed out on the second to last day of the course. This time, to main-
tain confidentiality and to ensure the students were able to give honest
responses, the forms were given to the class leader to hand out and collect.

Data Analysis

For Questionnaire A question 1 (Why did you choose this course?) was
analysed according to motivational type, that is, according to whether the
student’s responses appeared to indicate intrinsic, extrinsic, instrumental,
or integrative motivation (Deci and Ryan 1980; Gardner and Lambert
1959).
Responses to Questionnaire A question 2 (What do you want to get
out of this course?) were examined according to the type of need which
the student expressed.
The number of responses to each rating of Questionnaire B question
1 (regarding level of satisfaction with the course) were counted and aver-
aged overall.
For Questionnaire B, questions 2 (What were the positive aspects of
the Business English course?) and 3 (Were there any negative aspects of
the Business English course?), the responses were examined qualitatively
for any positive or negative perceptions of the Business English course.

Results

Motivation: Why Did Students Choose an ESP Course?

When categorizing the responses the students gave to this question

according to motivational type, it became obvious that by far the major-
ity (10 out of the 12 students) had an essentially instrumental reason
(Gardner and Lambert 1959) for choosing the Business English class. In
other words, they chose this class in order to use it to achieve some other
goal. These goals included:
 8  Example Studies 
   195

• Improving their English so that they could pass an exam (such as

International English Language Testing System (IELTS))
• Improving their English so that they could qualify to enter another
course (such as an Master of Business Administration (MBA))
• Improving their English and knowledge of business terminology so
that they would be better able to do business either in New Zealand or
when they returned to their home countries
• Qualifying for a better job with accompanying higher pay and higher
status
• Being able to provide a better life for their families

The two students who did not give one of the above reasons said they
had chosen the Business English class because GE was “boring” and they
wanted a change. Since this somewhat negative motivation might be
considered as an avoidance tool, perhaps it is more instrumental than
anything else.
Reasons which might be considered essentially extrinsic (Deci and
Ryan 1980) were given by five students, that is, they felt they were taking
the course because of some external influence. These external pressures
included especially:

• Parents (who were often the ones paying for the course, and who, in
some cases, were offering various incentives such as holidays, cars, etc.)
• Bosses (who wanted employees who could perform more effectively in
the international marketplace and who in some cases were offering
promotion and/or pay rises upon successful completion of the course)

Integrative reasons (Gardner and Lambert 1959) were noted by four

members of the class. These responses mostly involved:

• Being able to communicate with fellow workers

Interestingly, none of the students gave a reason which might be con-

sidered intrinsic (Deci and Ryan 1980). That is, nobody said they were
taking the course because they thought it was a good thing to do or they
were interested in it for its own sake. Several of the students gave more
196  Developing Language Teacher Autonomy through Action Research

than one reason for taking the Business English course, which underlines
the idea that motivation is frequently not a simple phenomenon, but that
individuals often have multiple reasons for what they do.

Needs Analysis: How Do ESP Students View Their Needs?

As can be seen from the list of instrumental motivations noted above,

most of the reasons mentioned by the students revolved in one way or
another around improving their language ability. In other words, for
them, the emphasis in “Business English” was on the “English”. The
“Business” was essentially the topic which provided (in most cases) the
context for improving their grasp of the language. The language which
was noted as requiring improvement included:

• Speaking skills, especially the ability to speak to clients or workmates

• Listening skills, especially being able to understand workmates, tele-
phone conversations, or financial news on TV
• Writing skills, especially being able to write business letters without
making embarrassing mistakes
• Reading skills, especially being able to read the business/financial sec-
tions of newspapers or magazines
• Business terminology
• Idiomatic expressions, especially of the colloquial kind used with
workmates

In other words, as can be seen from the list above, students saw their
main needs as being skills related, though they also expressed the need to
develop a wider knowledge of business and idiomatic vocabulary.

Course Evaluation: Do Students Think Their ESP Course Met Their

Needs?

When the results of the course satisfaction scale (where 5 = strongly agree
and 1 = strongly disagree that the course met their needs) were analysed
the results were as follows:
 8  Example Studies 
   197

• 5 (strongly agree) = 3 responses

• 4 (agree) = 7 responses
• 3 (neutral) = 2 responses
• 2 (disagree) = 0 responses
• 1 (strongly disagree) = 0 responses

In other words, the responses were all at the medium to upper end of
the scale, with most responses (n = 7) being in the “agree” range.

What Were the Positive Aspects of the ESP Course?

A number of positive aspects of the Business English experience were

mentioned on the survey forms, in particular:

• The Business English class was more interesting than GE. Several made
special mention of the trips to different businesses that we took as a
class, or the visits by businessmen to the class as being especially
interesting.
• The class was more motivating because it emphasized “useful” lan-
guage, so they could see a long-term purpose for what they were doing
• Students could see a relationship between the subject matter of the
Business English class and their eventual goals in life
• Because the class was separate from GE classes and stayed together for
12 weeks, the students developed a bond with each other and with
their teacher (myself ) which was often absent from GE classes

Were There Any Negative Aspects of the Course?

Although the feedback regarding the class was mainly positive, some neg-
ative points were raised:

• One student commented that the Business English class was not as
different from GE classes as hoped—it was really just GE with busi-
ness topics.
198  Developing Language Teacher Autonomy through Action Research

• A couple expressed disappointment that their English had not

improved more than it had.
• Several others commented that they had enjoyed the Business English
class very much, but now the course was finished, and they were going
to find it quite hard to go back to “boring” General English.

Discussion

The study has produced some interesting findings related to ESP courses
such as the Business English course involved here. These findings relate
especially to the nature of students’ motivation on such courses, to the
nature of their needs, to their level of satisfaction, and to some of the
positive and negative aspects of such courses.
Motivation, of course, is a key factor in successful learning, and it is
therefore important to keep clearly in mind what it is that motivates stu-
dents to take a course such as Business English. It is difficult to disagree
that intrinsic motivation is the best—it would seem intuitively obvious
that students who want to do something are the most likely to have the
drive to succeed. Most of the students in this study, however, seemed to
want to use it as a tool to achieve something else (instrumental), while
several others said they were taking the course because other people
wanted them to do it (extrinsic) rather than because of a strong desire on
their own part. For others, the main motivation was integrative—they
wanted to be able to communicate with those around them. However,
the fact that students did not express strong intrinsic motivation did not
stop this from being a delightful class.
The needs analysis clearly put the emphasis on the “English” rather
than on the “Business”. Students were actually very clear and realistic
about what they needed (mainly skills and vocabulary), and they wanted
to finish the course having improved in these areas. They did not want
or expect to finish the course as expert businesspeople. Some of them, in
fact, were already successful businesspeople either in New Zealand or in
their own countries. Others planned to go on to do higher degrees such
as an MBA in order to learn more about business. What they wanted
from the course was to be able to operate in English.
 8  Example Studies 
   199

It was satisfying to find that most of the students (10 out of the 12
or 83%) agreed that the course had met their needs. Satisfaction with
the Business English class which provided the context for this study was
generally high, and most of the students said they had enjoyed the class
and found it useful. They appreciated the relevance of the course to their
eventual goals and also the bonding that was possible with classmates
which was much more over a 12-week period than in GE classes where
students were constantly arriving and leaving. Of those who were less
enthusiastic, probably the two who had taken the class because GE was
“boring” appeared to be the least satisfied. Although it was not possible to
positively identify individuals from the anonymous end-of-course survey,
these two, in fact, continued to have relatively poor attendance records,
often arrived late when they did come to class, were often “sick”, and
were usually at or near the bottom of the class in tests. Perhaps this could
be taken as a caution to those who might see an ESP class as a quick fix
for language learning difficulties. The reality is that if students are not
prepared to work, it doesn’t matter what class they are in, they will not
learn!

Implications for the Teaching/Learning Situation

In response to the findings noted above, there are several suggestions we

might make for the teaching/learning situation:

1. These findings underline yet again the importance of motivation in

successful learning. Therefore it is obviously necessary for teachers and
teaching institutions to strive to provide students with a motivating
environment, high interest materials, and supportive staff.
2. Having said this, it must be realistically borne in mind that where
students are unmotivated, it is not always the fault of the teacher or
the school. This point is demonstrated by the two “bored” students in
this class: they chose the class because they were “bored” and they
continued to apply minimal effort to their studies. In fact, they did
not want to be business people, or even to study English at all, so they
did not have the kind of positive motivation the others had, and in the
200  Developing Language Teacher Autonomy through Action Research

end they got no more out of this class than out of any other. So we
need to be careful about expecting miracles from ESP classes.
Ultimately, success is up to the student, and unmotivated students
will not succeed no matter what the nature of their course.
3. This study shows very clearly the importance of remembering the “E”
in “ESP”. Students tend to take such courses essentially to improve
their English. The special purpose (be it business or whatever) pro-
vides the context and may contribute to motivation, but essentially
such courses are about developing the language, and this needs to be
clearly borne in mind when designing and conducting such courses.
4. The concern that some of the students felt at needing to reintegrate
into the GE stream at the end of the Business English course described
in this study points to another factor for course scheduling: it is better,
if possible, to offer such courses at the end of a students’ time at a
school, as otherwise they may experience problems with being assimi-
lated back into the mainstream classes.

Suggestions for Ongoing Research

This study suggests a number of areas requiring further research. In

particular:

1. The whole area of motivation in ESP needs further exploration.

Especially important are the questions of what motivates students,
what keeps them motivated, what is the role of intrinsic motivation,
perhaps as an underlying driving force, and how do multiple motiva-
tions (such as instrumental and extrinsic) interact with each other?
2. Student needs is another big area awaiting further clarification. We
need to know what, exactly, do students need from ESP courses, how
do these needs differ from other courses, if at all, and how, precisely,
can needs best be met in ESP courses?
3. What can we do about unmotivated students? This question is, of
course, not unique to ESP, but it applies to ESP as much as to any
other course. The expectation that ESP courses might magically pro-
 8  Example Studies 
   201

vide unmotivated students with the will to work and succeed at what
they are doing is, unfortunately, probably rather naïve.

Conclusion

ESP has had a long and at times controversial history. Opinions have
varied all the way from Widdowson’s (1983) comment that “all language
courses are designed to a specification and in this sense, all can be said
to be directed at specific purposes” (p. 12), which would seem to suggest
that ESP is not very different from any other language course. Johns and
Dudley-Evans (1991), however, argue that ESP is “international in scope
[and] specific in purpose” (p. 297).
The current study, however, seems to have produced evidence that
such courses can satisfy a need for some students in as far as they seem
to help students focus on a goal beyond the day-to-day business of learn-
ing language, thereby providing motivation. More research needs to be
done, however, on more careful analysis of the nature of the motivation
involved, on the nature of the needs and how these can be met, and how
to deal with unmotivated students.

References

Adelman, C., Jenkins, D., & Kemmis, S. (1976). Re-thinking case study:
Notes from the second Cambridge conference. Cambridge Journal of
Education, 6, 139–150.
Brinton, D., Snow, M., & Wesche, M. (1989). Content-based second
language instruction. Michigan: University of Michigan Press.
Cervi, D. (1993, November). ESP ousts general English. EFL Gazette,
London, p. 14.
Deci, E., & Ryan, R. (1980). The empirical exploration of intrinsic
motivational processes. In L. Berkowitz (Ed.), Advances in experimental
social psychology (Vol. 13, pp. 39–80). New York: Academic Press.
Gardner, R., & Lambert, W. (1959). Motivational variables in second
language acquisition. Canadian Journal of Psychology, 13, 266–272.
202  Developing Language Teacher Autonomy through Action Research

Hutchinson, T., & Waters, A. (1987). English for specific purposes: A

learning-centred approach. Cambridge: Cambridge University Press.
Johns, A., & Dudley-Evans, T. (1991). English for specific purposes:
International in scope, specific in purpose. TESOL Quarterly, 25/2,
297–314.
Johnson, D. (1992). Approaches to research in second language learning.
New York: Longman.
Jones, L., & Alexander, R. (1989). International business English.
Cambridge: Cambridge University Press.
Nunan, D. (1992). Research methods in language learning. Cambridge:
Cambridge University Press.
Oxford, R. (1996). Employing a questionnaire to assess the use of lan-
guage learning strategies. Applied Language Learning, 7/1&2, 25–45.
Pinto Da Silva, C. (1993). ESP: Back to methodology. Forum, 31/2,
40–41.
Richards, J., Platt, J., & Platt, H. (1992). Longman dictionary of lan-
guage teaching and applied linguistics. London: Longman.
Robinson, P. (1991). ESP today, a practitioner’s guide. New  York/
London: Prentice Hall.
Schmidt, M. (1981). Needs assessment in English for specific purposes:
The case study. In L. Selinker, E. Tarone, & V. Hanzeli (Eds.), English for
academic and technical purposes (pp. 199–210). Rowley: Newbury House.
Strevens, R. (1988). ESP after 20 years: A re-appraisal. In M. Tickoo
(Ed.), ESP: State of the art (pp.  1–13). Singapore: SEAMO Regional
Language Centre.
Widdowson, H. (1974). An approach to the teaching of scientific
English discourse. RELC Journal, 5(1) 27–40.
Widdowson, H. (1983). Learning purpose and language use. Oxford:
Oxford University Press.

Appendix A: Motivation and Needs Analysis Form

Please answer the following questions:

Why did you choose this course?
What do you want to get out of the course?
 8  Example Studies 
   203

Appendix B: Course Evaluation Form

Please respond to the following statement according to whether you

5 = strongly agree 4 = agree 3 = neutral 2 = disagree 1 = strongly

disagree.
That the Business English course met your needs well.
_____________________________________________________
___________________.
What were the positive aspects of the Business English course?
_____________________________________________________
___________________.
Were there any negative aspects of the Business English course?

A Questionnaire-Based Study
Gender and Strategies in the Language Classroom

Yusuf Çeribaş

Abstract
This study investigated the role played by gender in the language class-
room, especially in relation to the role played by language learning strate-
gies. The study included 139 students at a University Preparation School
in Istanbul, Turkey, of whom 71 were male and 68 were female. It was
found that these students reported quite high levels of strategy use, but
that only two strategy items were significantly correlated with achieve-
ment. There were no significant differences according to gender for either
strategy use or achievement.

Introduction

Gender is considered to be one of the factors affecting language acquisi-

tion (e.g. Sunderland 1998), although few studies have been conducted
204  Developing Language Teacher Autonomy through Action Research

to explore its impact and to suggest ways to successfully adjust teaching

to accommodate gender differences in the classroom. Language learn-
ing strategies are also believed to have a strong effect on language acqui-
sition (e.g. Cohen 2011; Griffiths 2013; Oxford 2011). However, the
relationships between and among gender, strategies, and achievement in
language learning have not been well researched.

Literature Review

There are a number of studies which have concluded that females are
more adept at learning language than males. A study by Boyle (1987)
found that among 490 Chinese university students in Hong Kong, the
mean score of female students on second language proficiency tests was
higher than that of males. Ellis (1994) also reported that, according to
research, females tend to be more capable of learning foreign languages
than males. Statistics from high-stake exams such as Test of English as a
Foreign Language (TOEFL) and International English Language Testing
System (IELTS) seem to confirm the difference in success rates between
male and female learners: based on the data, female test takers scored
slightly higher than male test takers on the IELTS Academic Test (James
2010). A study by Park and French (2013) had a similar finding: they
studied 948 university students who learned English in Korea, and the
female learners in their study received a higher final grade than males.
The use of strategies to assist learners to develop language has long
been acknowledged (e.g. Rubin 1975), although the strategy concept
has remained controversial. According to a recent definition by Griffiths
(2015), language learning strategies are “actions chosen by learners (either
deliberately or automatically) for the purpose of learning or regulating
the learning of language” (p. 426). Intuitively, it would be reasonable to
expect faster progress from learners who are more capable of regulating
their own learning. In a study by Green and Oxford (1995) which had
374 students as subjects at the University of Puerto Rico, a significant
relationship between strategy use and proficiency was reported. When
Dreyer and Oxford (1996) conducted a study on ESL proficiency in
South Africa, they found a significantly positive correlation between suc-
cessful TOEFL scores and the frequency of strategy use. Park (1997) also
 8  Example Studies 
   205

set out to investigate the relationship between language learning strate-

gies and TL proficiency, using the SILL (Strategy Inventory for Language
Learning, Oxford 1990) and the TOEFL as instruments: all six categories
of language learning strategies were found to be significantly correlated
with the TOEFL scores. Likewise, Griffiths (2013) reported a study with
348 students at a language school in New Zealand which noted similar
findings.
The results of the studies in the language learning strategy literature sug-
gest that males and females deal with language learning in different ways
(e.g. Oxford and Nyikos 1989). A majority of studies has found that females
use strategies more frequently than males, and are often more successful.
For instance, in Green and Oxford’s (1995) study which was conducted
at the University of Puerto Rico with 374 students, it was concluded that
females used language learning strategies significantly more frequently
than men and achieved significantly higher levels of language proficiency.
According to a study reported by Griffiths (2013) however, although the
females in the study scored slightly higher grades and used language learn-
ing strategies slightly more frequently, the differences were not significant.

Research Questions

This study aimed to explore the following questions:

1. What is the reported level of strategy use among the students in this
context, both overall and according to gender?
2. What is the relationship between language learning strategies and
achievement in the language classroom?
3. What is the relationship between gender and achievement?

Research Context

Setting

This study was carried out at the Preparatory School of a private univer-
sity in Istanbul, Turkey.
206  Developing Language Teacher Autonomy through Action Research

Since there were eight classes, the teachers collaborated to help collect,
match, and analyse the data and then share the results.

Participants

In total, there were 164 students who took part in this study. They were
all between 18 and 20 years old preparing to enter differing departments
and faculties at the university. Due to the lack of name information on
some forms and the presence of forms with invalid responses, 139 stu-
dents’ responses were included, all of which could be matched with the
proficiency exam scores. Of the 139 valid responses, 71 belonged to male
students and 68 belonged to females.

Data Collection

Permission for the study was obtained from the university authorities.
Students were informed about the nature of the study and asked to sign
that they consented to the use of the data for research and possible pub-
lication purposes. The English Language Learning Strategy Inventory
(ELLSI 2015), an updated version of the one used in Griffiths (2003),
was used to collect the strategy information (see Appendix). The students
were asked to rate each strategy item in the questionnaire from 1 to 5
according to how frequently they used it. The ELLSI for students was
administered by the teachers of the English Preparatory School. A class-
room time of around 15 minutes was allocated for the administration
of the surveys which were collected by the teachers immediately after
they were completed in the classrooms and submitted to the researcher
for data analysis. In addition, the students’ test scores were gathered by
means of routine testing procedures.

Data Analysis

The software package “Statistical Package for Social Sciences (SPSS)” was
used to find:
 8  Example Studies 
   207

(i) Normality of distribution

(ii) Reliability coefficient
(iii) Medians
(iv) Correlation coefficients (Spearman)
(v) Differences (Mann–Whitney U)
(vi) Percentages

Results

To decide whether the data set requires parametric or nonparamet-

ric tests, the Kolmogorov–Smirnov (K-S) test was conducted for the
ELLSI. According to the results of the K-S test, the significance value for
the items in the survey was below .05, and this means that there is no
normal distribution and therefore nonparametric tests were found more
appropriate for this study.
An alpha test of reliability showed a Cronbach’s alpha (α) coefficient of
.857, which is quite a high level of reliability (Dörnyei 2007).

The Reported Level of Strategy Use Among the Students Both Overall

and According to Gender

Table 8.3 lists the median frequency of use for each strategy, for males,
females, and overall.

The Relationship Between Language Learning Strategies

and Achievement

When the achievement and strategy use of participating students were

correlated, only two significant correlations were found using Spearman’s
rho. As can be seen in Table 8.4, there is a significant positive correlation
between pass–fail grade and learning from the teacher (p < .05), and there
is a significant negative correlation (p < .01) between pass–fail grade and
writing a diary in English
208  Developing Language Teacher Autonomy through Action Research

Table 8.3  Strategy frequency medians for males, females, and overall
Item Males’ Females’ Overall
No. Statements median median median
1 Doing homework 3 3 3
2 Learning from the teacher 4 4 4
3 Reading books in English 2 3 3
4 Writing a diary in English 2 2 2
5 Watching TV in English 4 4 4
6 Revising regularly 3 3 3
7 Listening to songs in English 4 4.5 4
8 Using language learning games 3 3 3
9 Writing letters in English 2 2 2
10 Listening to the radio in English 2 2 2
11 Talking to other students in 3 3 3
English
12 Using a dictionary 4 4 4
13 Reading newspapers in English 2 2 2
14 Studying English grammar 3 3 3
15 Consciously learning new 4 4 4
vocabulary
16 Keeping a language learning 4 3.5 4
notebook
17 Talking to native speakers of 3 2 3
English
18 Learning by APPs on the mobile 3 3 3
phone
19 Controlling schedules so that 3 2.5 3
study is done
20 Taking online English courses 2 2 2
21 Not worrying about mistakes 3 3 3
22 Trying to think in English 4 4 4
23 Listening to native speakers of 4 3 4
English
24 Learning from mistakes 4 4 4
25 Spending a lot of time studying 3 3 3
English
26 Making friends with native 3 2 3
speakers
27 Watching movies in English 4 4 4
28 Learning about the culture of 3 3 3
English speakers
 8  Example Studies 
   209

Table 8.4  Significant correlations between pass–fail grades and strategy items 
Learning from the teacher Writing a diary in English
Pass–fail .196* −.326**
* = significant at the p < .05 level
** = significant at the p < .01 level

The Relationship Between Gender and Achievement

Among these students, 90 passed and 49 failed. (Minimum score to pass

is 60 out of 100.) Of the 68 female students who took the proficiency
exam, 42 (61.8%) passed and 26 (38.2%) failed. Of the 71 male stu-
dents who took the proficiency exam, 48 (67.6%) passed and 23 (32.4%)
failed. The figures seem to favour boys over girls even though the differ-
ence in success rate according to gender is not significant.

Discussion

The strategies which had the highest rating (4 = often) are learning from
the teacher, watching TV in English, listening to songs in English, using
a dictionary, consciously learning new vocabulary, keeping a language
learning notebook, trying to think in English, listening to native speak-
ers of English, learning from mistakes, and watching movies in English.
Even though students tend to use multimedia to improve their language
skills, they still see their teachers as a main resource. The findings suggest
that they attach importance to vocabulary learning since they reported a
frequent use of dictionaries and notebooks. The findings also show that
they see feedback as crucial for language learning and they tend to appre-
ciate the didactic side of making mistakes.
The strategies which had the lowest rating (2 = not very often) are
writing a diary in English, writing letters in English, listening to the radio
in English, reading newspapers in English, and taking online English
courses. Students in this study probably see diaries and letters as out-
dated. They may be more inclined to use email and to write blogs. The
students are also honest about foreign radios and newspapers since they
are not easily accessible in a Turkish environment. Probably due to the
fact that these students have chosen the traditional way of learning a
210  Developing Language Teacher Autonomy through Action Research

language, which is by going to school five times a week, they may not be
interested in the online courses available.
Around 65% of the students who sat the exam passed it, which is a
very good figure to help us understand the academic profile of the stu-
dents in this study. The minimum grade required to pass is 60%, and a
majority of the students were able to get at least 60% of the tasks in the
exam right. When the researcher looked for correlations between stu-
dents’ achievement and strategies, however, he found only two signifi-
cant correlations: a positive significant correlation between achievement
(pass–fail grade) and learning from the teacher (statement no. 2 in the
ELLSI) and a significant negative correlation between achievement and
writing a dairy in English. From these results, we might conclude that,
although the students report using (median = 4) many of the strategies (n
= 10) often, there does not seem to be a strong relationship between most
of the items and success.
If we look at differences in strategy use according to gender, we can see
from Table 8.1 that males report using ten strategy items “often”, whereas
females report using only seven items at this rate of frequency. At the
same time, females are alone in giving a rating of 4.5 to Item 7 Listening
to songs in English. But none of these differences were significant, leading
to the conclusion that males and females do not differ widely according
to strategy preferences.
To explore the impact of gender on achievement, we can take a closer
look the success rates of females and males. While the percentage of suc-
cessful learners is 62% for female learners, it is around 68% for male
learners. Though both groups seem to have done well in the exam, the
figures are slightly better for males. Although this difference did not
prove to be significant, it is, nonetheless, somewhat different from most
previous studies which have tended to favour females.

Pedagogical Implications

One surprising finding of this study was that taking online courses was
rated very low by the participating students. In this technological age,
when everyone seems to be talking about and using hi-tech gadgets, this
 8  Example Studies 
   211

is hard to explain. Is it because students don’t know about these courses,

or is it that, in spite of the ubiquity of technology, people actually do
prefer interacting with another human? If the former, maybe teachers
should be prepared to provide their students with the relevant informa-
tion to enable them to use these courses effectively. If the latter, teachers
should, perhaps, be relieved and happy: they are not about to be replaced
by a machine any time soon. This interpretation also resonates with the
high median rating given by all three groups (male, female, and overall)
to Item 4: Learning from the teacher.
Considering the findings of this study, it can be argued that teachers
need to be more aware of the presence of gender as an individual dif-
ference in their classes. Gender may have an impact on other aspects of
learning in addition to strategy use and achievement in language l­ earning.
Therefore, teachers ought to try to understand the various factors includ-
ing gender that affect their perceptions, learners’ use of language learn-
ing strategies, students’ overall success, and classroom dynamics in the
foreign language classroom. It would be sensible to suggest that the bet-
ter individual factors within the language classroom are understood, the
more effective teaching can be.

Limitations and Suggestions for Further Research

There are several limitations of this study in terms of data collection and
sampling. To begin with, the study relied heavily on surveys and this can
result in a narrow perspective in results. More triangulation by means
of a variety of methods might be useful. Structured or semi-structured
interviews could be used and think-aloud protocols with several students
would add to the validity of this study.
Another limitation is related to the sample size and the context. A
larger sample could have been chosen for more reliable results. All the
students who participated in the study were studying at a foundation
university in Istanbul, Turkey. There were no students from state univer-
sities, high schools, primary schools, and elementary schools. The survey
could be administered in several other schools, and in different locations
to see if results are similar.
212  Developing Language Teacher Autonomy through Action Research

Furthermore, convenience sampling was used for practicality and to

avoid administrative issues. However, random sampling would yield
more valid and reliable results.
Considering the limitations this study has, the relationship between
and among gender, strategy use, and achievement needs to be researched
with larger samples and in multiple contexts so that the questions can be
explored in more detail and studied more thoroughly. Researchers who
are interested in this particular subject may also utilize different methods
such as interview, observation and a variety of other instruments.

Conclusion

Several interesting findings have emerged from this study. In particu-

lar, it appears that students are already active strategy users, in that they
report using ten strategies “often”, but only five “not often”. Although
some studies have reported a significant relationship between strategies
and successful learning (e.g. Green and Oxford 1995; Griffiths 2013),
in the case of this study, only two strategies were found to be signifi-
cantly related: one positive (Item 2 Learning from the teacher) and one
negative (Item 4 Writing a diary in English). Furthermore, there was no
significant difference between males and females either for strategy use or
for achievement. Although females report using more strategies “often”,
67.6% of the males passed the English proficiency exam compared with
only 61.8% of the females. In other words, we can conclude that, accord-
ing to the results of this study in this context, there is remarkable uni-
formity of results, with no significant differences in level of achievement
according to either strategy use or gender.

References

Boyle, J. (1987). Sex differences in listening vocabulary. Language

Learning, 37, 273–284.
Cohen, A. (2011). Strategies in learning and using a second language
(2nd ed.). London: Longman.
 8  Example Studies 
   213

Dörnyei, Z. (2007). Research methods in applied linguistics. Oxford:

Oxford University Press.
Ellis, R. (1994). The study of second language acquisition. Oxford:
Oxford University Press.
Green, J., & Oxford, R. (1995). A closer look at learning strategies,
L2 proficiency, and gender. TESOL Quarterly, 29(2), 261. http://doi.
org/10.2307/3587625
Griffiths, C. (2003). Language learning strategy use and proficiency.
http://hdl.handle.net/2292/9
Griffiths, C. (2013). The strategy factor in successful language learning.
Bristol: Multilingual Matters.
Griffiths, C. (2015). What have we learnt from good language learn-
ers? ELTJ, 69/4, 425–433.
James, C. (2010). Do language proficiency test scores differ by gen-
der?. TESOL Quarterly, 44(2), 387–398.
Oxford R. (1990). Language learning strategies: What every teacher
should know. New York: Newbury House.
Oxford R. (2011). Teaching and researching language learning strategies.
Harlow: Pearson Longman.
Oxford, R., & Nyikos, M. (1989). Variables affecting choice of lan-
guage learning strategies by university students. The Modern Language
Journal, 73(3). 291–300.
Park, G., & French, B. (2013). Gender differences in the foreign lan-
guage classroom anxiety scale. System, 41(2), 462–471.
Rubin, J. (1975). What the ‘good language learner’ can teach us.
TESOL Quarterly, 9(1), 41–51.
Sunderland, J. (1998). Girls being quiet: A problem for foreign lan-
guage classrooms? Language Teaching Research, 2(1), 48–82.

 ppendix: English Language Learning Strategy Inventory

A
(ELLSI): 2015

Name: ___________________________M/F
214  Developing Language Teacher Autonomy through Action Research

II. Dear student: please read the following list of language learning

strategies. Please mark each one according to how often you use it
5=very often  4=often  3=sometimes  2=not often  1=never
Item Strategy Rating Comment
1 Doing homework
2 Learning from the teacher
3 Reading books in English
4 Writing a diary in English
5 Watching TV in English
6 Revising regularly
7 Listening to songs in English
8 Using language learning games
9 Writing letters in English
10 Listening to the radio in English
11 Talking to other students in English
12 Using a dictionary
13 Reading newspapers in English
14 Studying English grammar
15 Consciously learning new vocabulary
16 Keeping a language learning notebook
17 Talking to native speakers of English
18 Learning by APPs on the mobile phone
19 Controlling schedules so that English study is done
20 Taking online English courses
21 Not worrying about mistakes
22 Trying to think in English
23 Listening to native speakers of English
24 Learning from mistakes
25 Spending a lot of time studying English
26 Making friends with native speakers
27 Watching movies in English
28 Learning about the culture of English speakers
Any other strategies you are aware of using

I consent to these data being used for research and/or

publication: (signature)
 8  Example Studies 
   215

Questions to Consider
1. In your environment, what do you see as the dis/advantages of

(a) Quasi-experiments
(b) Case studies
(c) Questionnaire studies

2. Do you agree with Koç) that “Extensive reading should not be taken
only as a free time activity, because it really is a serious, useful learning
experience which should be valued and introduced as an integral aspect
of the classroom environment”?
3. Do you agree with Griffiths that ESP courses “can satisfy a need for some
students in as far as they seem to help students focus on a goal beyond
the day-to-day business of learning language, thereby providing
motivation”?.
4. Do you think that you would come to the same conclusion as Çeribaş
(that “there is remarkable uniformity of results, with no significant dif-
ferences in level of achievement according to either strategy use or gen-
der” in your environment?
 9
Presenting and Writing Up Action
Research

Although many action researchers may want to conduct their studies

purely for their own interest and satisfaction and to inform their own
classroom practices, many others wish to present their results to a wider
audience. Indeed, presentation can be an important stage of any research
project, since it provides the opportunity to identify new findings, to
make them available to others, to get feedback, to obtain new ideas, and
perhaps to progress professionally. There are a number of ways that results
can be reported:

• In the teaching environment, such as at a staff meeting. A sound

research culture is essential if this kind of activity is to be successful.
• At a conference, for example, by means of a paper, a poster, a workshop.
• On websites, for example, Academia.edu (www.academia.edu), Research
gate (www.researchgate.net).
• On a blog, vlog, or website.
• Writing an assignment or thesis.
• Writing a chapter in an edited book.
• Writing a book.
• Submitting it to a journal.

© The Author(s) 2017 217

K. Dikilitaş, C. Griffiths, Developing Language Teacher Autonomy
through Action Research, DOI 10.1007/978-3-319-50739-2_9
218  Developing Language Teacher Autonomy through Action Research

In-House Presentations
Many teachers may be quite content to do a relatively low-key presenta-
tion of what they have done and what they have found to a small group
of interested colleagues in their own teaching environment. This can be
a good way of practising presentation skills and gaining confidence for
larger-scale events if this is what is wanted.
By doing this, researchers can develop awareness of the issues other
teacher researchers have problematized. In-house presentation can
also contribute to school development since common issues could be
discussed from different points of view. For example, if a researcher
has investigated an issue that other teachers have also been puzzled
by, then such research could initiate contextualized discussions for
development.

Conference Presentations
Others may enjoy the challenge and the opportunities of presenting their
research at a conference. There are numerous organizations which pro-
vide opportunities for this.

Where?

IATEFL (International Association of Teachers of English as a Foreign

Language—UK). This is a very big international conference held annu-
ally at various centres around the UK, usually around April. It tends to
attract many of the “big names” in the ELT field, and to be practitioner-­
focused (rather than highly theoretical). In addition to the main con-
ference, Special Interest Groups (SIGs) attached to IATEFL (e.g. the
Research or the Teacher Education SIGs) often hold smaller conferences
in various locations around the world. The website is www.iatefl.org
TESOL (Teaching English to Speakers of Other Languages—
USA). This is another big international conference, also held annu-
 9  Presenting and Writing Up Action Research 
   219

ally around April at various locations in North America. TESOL also

tends to focus on teachers rather than on theoretical issues and also
attracts many well-­known speakers. Furthermore, there are also many
affiliates which hold their own conferences worldwide. The website is
www.tesol.org
AILA (International Applied Linguistics Association). This is another
big conference held every three years at various locations around the
world. As the “Applied Linguistics” in the name implies, it is more
theoretically oriented than IATEFL or TESOL.  The website is www.
aila.org
CLESOL (Community Languages and English for Speakers of Other
Languages—New Zealand). With a tendency to be practitioner-focused,
this is a biennial conference held at various locations around New
Zealand. The website is www.clesol.org.nz
AAAL (American Association of Applied Linguistics—USA). This is
an annual conference which tends towards the theoretical, and is usually
held around the same time and place as TESOL, meaning that it can be
difficult to get enough time away from work to attend both, but great if
this is not a problem. The website is www.aaal.org
EUROSLA (European Second Language Association—Europe). This
is another big annual conference, held at various locations in Europe. The
website is www.eurosla.org
GALA (Greek Applied Linguistics Association—Greece). Centred in
Thessaloniki in Greece, GALA holds conferences in various locations in
Greece. The website is www.enl.auth.gr/gala
AsiaTEFL (Asian Teachers of English as a Foreign Language) is an
association which specializes in Asian issues with the teaching of English.
Annual conferences are held in various locations around Asia. The web-
site is www.asiatefl.org
T-PLUS (Trainers’ Professional Learning and Unlimited Sharing)—
Turkey). T-PLUS holds an annual conference at various locations around
Turkey and beyond. The website is www.tplustr.org
In addition, there are numerous other organizations which hold con-
ferences all over the world.
220  Developing Language Teacher Autonomy through Action Research

How

There are numerous ways to participate in conferences:

• Poster presentation. This involves preparing a poster with the basic

information from the research presented visually. Typically, the confer-
ence organization will provide a time and a place to display the poster,
and there is the opportunity to stand beside the poster and explain it
to interested observers. The requirements for the poster (e.g. size) are
usually carefully specified by the conference, so be sure to check these
with the conference website. These days there are some useful online
programmes (e.g. “Glogster”, “Pictochart”) which help to produce
interesting, attractive, and professional-looking posters reasonably eas-
ily as per our example (Fig. 9.1).

Fig. 9.1  Sample poster (http://rukie.edu.glogster.com/fostering-speaking/

with Rukiye’s permission)
 9  Presenting and Writing Up Action Research 
   221

• Present a paper, either jointly with a colleague or individually. For

those who prefer a more academic approach, presenting research as a
paper can be a good way to achieve exposure. Presentation times are
typically 20–30  minutes. In order to produce an interesting paper,
some kind of visual material may be necessary. There are a range of
such programmes available these days (e.g. Prezi), but some good
results can be obtained with the more familiar PowerPoint, especially
if some more visual material such as photos or clip art are included to
break up what might otherwise be long and indigestible text.
• Present a workshop. Workshops are more practical than papers (which
tend to be more lecture-like) and involve more audience participation.
Also, typically, more time is allowed for a workshop than for a paper.
• Join a symposium or colloquium, which usually involves joining a
team of others talking on related aspects of a topic. These can be a
good idea, since they provide the support of the others, whereas stand-
ing and presenting alone can be a very intimidating experience.

Online Presentation
In this day and age it is impossible to ignore the impact of technology
and the Internet. Many people have their own blogs or websites (e.g.
www.carolgriffiths.net) and these can be a very useful way of making
information available for others. For those who do not wish to go to the
trouble and to spend the time involved in establishing and maintaining
such sites, there are sites such as www.academia.edu or www.researchgate.
net which can be used to notify others.

Self-Videotaping and Broadcasting
It is also possible to record videotapes while presenting the study with a
poster and then share it through social media such as Facebook, Twitter,
or YouTube. Such dissemination will help access a number of people who
are interested. Teachers can develop their confidence in presenting their
222  Developing Language Teacher Autonomy through Action Research

studies with several rehearsals and create an effective delivery. R

­ eal-­time
presentations can sometimes cause problems if the presenters are not
well-prepared. Vlogs—video blog—allows for supporting video with
text, images, links, and so on. This support can help teachers improve the
quality of their delivery.

Written Presentations
Doing action research (AR) and writing it up are two different processes in
that some may opt for not writing up the research for dissemination as they
might feel that doing it helped them understand the questions, puzzles, or
issues, and they are happy to leave it at that. However, just as they might
learn a great deal about the research focus just by doing the study, the
writing-up process can help to develop new ideas and also to develop writ-
ing skills as it is written up. Being (possibly) inexperienced researchers and
writers, action researchers (especially teachers whose first language is not
English) could face challenges in writing up their research for international
readers. However, without trying to write and submitting their work to get
feedback, they will never develop their writing skills.
There are different ways of starting writing up. Some prefer to write
simultaneously as they complete specific parts of their research. For
example, starting with reading relevant studies, they may try to collate
the literature review and identify points that could confirm and discon-
firm the research points. Others may prefer to do research first, and then
bring it together afterwards. Each strategy is fine as long as it feels com-
fortable. However, our recommendation would be taking thick notes
while engaged in doing the research so that the lived experiences are not
forgotten. For example, data collection procedures could be written in
situ as it happens since it could be hard to write retrospectively.
There are also different views as to whether AR should be written within
the conventions of academic research or through a genre that combines a
reader-friendly style (keep in mind that the AR audience consists mostly
of practitioners rather than academics). Some views support the idea that
AR should be written within its own conventions, a kind of language that
is not challenging to read. There are also others who think that AR should
 9  Presenting and Writing Up Action Research 
   223

be written in an academic way since research, if published and dissemi-

nated, needs to be written within the writing conventions of academics.
Most commonly, these will take the form of

• An assignment or thesis being carried out in fulfilment of a qualifica-

tion (e.g. MA or PhD). If engaged in this kind of activity, try to think
beyond the immediate need to submit the material for assessment and
to consider how it might be put to further use, such as presenting it at
a seminar or conference. Otherwise it is easy to end up with shelves
full of dusty assignments, which often contain the germs of good ideas,
and which, with a little more thought, might be used for more inter-
esting purposes.
• A chapter in an edited book. Many would-be authors are looking for
material, and this may present a good publication opportunity.
Otherwise, journals often include calls for such material.
• A book. This might be a monograph (written by a single author):
many theses contain enough material to be turned into a book, per-
haps with some rewriting and extra material added. Or teaming up
and working with co-authors helps to spread the load (as long as every-
one is pulling their weight!). Or acting as editor and compiling various
sections written by others can be an option.
• An article in a newsletter, magazine, or journal. For those thinking of
going down this road, there are several things to consider, especially if
aiming for a good journal:
–– Title: The title is the point of entry for the reader of an article. It
needs to catch the reader’s attention, and it should not be too long
(many journals have a maximum of maybe ten words). Short titles
are often the most memorable:

• Interlanguage (Selinker 1972)

• Culture’s Consequences (Hofstede 1980)
• The Input Hypothesis (Krashen 1985)

All of these titles have echoed down the years and had a major impact
on thinking. Linguistic techniques may also help to catch attention, such
as the well-worn journalistic technique of alliteration:
224  Developing Language Teacher Autonomy through Action Research

Lessons from good Language Learners (Griffiths 2008)

Appealing for an emotional reaction may also be effective:
The more I learned, the less I found myself (Bian 2009)
–– Abstract: The abstract should give a concise idea of what the article
is about. A journal will probably give a word limit, which should be
maintained as closely as possible. Generally speaking, it should not
include references. But it should contain the main findings of the
research so that it can be quoted if necessary. Sometimes this is all
an author who wants to cite the work has access to, since most jour-
nals publish the abstracts online, but a subscription is often required
for the whole article. The abstract should probably be the last thing
written, because ideas change as the study progresses. What is in the
abstract should match what is in the article.
–– Keywords: The purpose of keywords is so that key topics will appear
on a database for someone looking for this subject. The title will
already be on the search engine so it is not necessary to include
words in the title. Good keywords are an important aid to citability,
so it is worth spending a bit of time thinking carefully about the
topics the article deals with and which might be of interest to some-
one researching a related topic.
–– Introduction: The introduction is generally reasonably brief, and
its job is to provide background, perhaps in the form of a rationale
for the study, and road map to explain what the article plans to
cover. What is it that the study is trying to achieve? The introduc-
tion should highlight this very explicitly so that readers can under-
stand the research aim. When the article is finished, it should be
checked against the introduction to ensure it has covered the
stated areas.
–– Literature review: The literature review should do exactly what it
says: it should review what has already been written on the sub-
ject. Most journals have rules regarding their literature, so it is
essential to check with these and to follow the requirements. Some
journals allow only limited literature reviews: ELT Journal, for
instance, allows only 15 references, and carefully limits in-text ref-
erences as well, which can be difficult to keep to. Other journals
 9  Presenting and Writing Up Action Research 
   225

allow lengthy reference lists and also permit multiple in-text refer-
encing. Referencing policy may well contribute to the decision
regarding which journal is most suitable for the article. In addi-
tion, and perhaps most importantly, the literature review should
look for gaps, that is, what has not been researched so far, so that
any new research is contributing something new and not merely
reinventing the wheel. To avoid long and loaded paragraphs, lists,
charts, or tables can be used to bring together a combination of
views concisely.
An example literature review by Merve Babiker, Sezen Savaş and
Gamze Taşlı (published in Dikilitaş, Wyatt, Hanks and Bullock, 2016)

Online Interaction and Collaborative Writing

Research on student interaction and collaborative writing has often been

viewed through what Ellis (1997) frames as two perspectives: the inter-
action hypothesis and the sociocultural theory. The interaction hypothesis
comes from the original notion that face-to-face communications offer
substantial benefits to learners by focusing their attention to both spo-
ken and written language, especially in the midst of communicative
mishaps (Blake 2008). The sociocultural perspective is adapted from the
Vygotskyan view of language learning through interactions via broad
social and cultural contexts. It centres upon the value of aid provided by
the interlocutor in order to achieve negotiations of meaning in language,
promoting a learner’s Zone of Proximal Development (ZPD). Vygotsky
(1978) explains ZPD in terms of the promotion of collaborative learning
through social interactions of both student-to-teacher and student-to-stu-
dent as “ the distance between the actual development level as determined
by independent problem solving under adult guidance or in collaboration
with more capable peers” (cited in Li and Zhu 2013, p. 86).
In the studies presented, the various benefits of collaboration between
learners of second languages have been debated. (Bruce et  al. 1993).
Context, tools, as well as the atmosphere in which learning takes place
can all have a crucial role in the mediation of collaborative learning.
Kessler (2009) describes that via collaboration, students’ exposure to
226  Developing Language Teacher Autonomy through Action Research

external comprehensible input is enhanced (Vygotsky 1962), produc-

tion of high-quality output is supported (Oxford 1997), and efficient
linguistic feedback can be provided for all participants (Vygotsky 1978).
Therefore, the act of collaborative writing can be enhanced through the
vast opportunities that technology can provide.
Asynchronous and synchronous communication
The issue of collaboration in the language learning process has long been a
topic of research. As Blake (2008) highlights, talented classroom teachers
look for ways to make their students take part in collaborative interac-
tions. He further adds that “teachers can create the same opportunities for
interactions within the context of computer-mediated communication
(CMC), whether in real time (synchronous computer-mediated com-
munication, SCMC) or deferred time (asynchronous computer-mediated
communication, ACMC)”. Asynchronous communication takes place
outside the real time. Email, electronic mailing lists, online discussion
boards, wikis and blogs are tools for asynchronous communication. Using
these tools as a part of classroom instruction provides many opportunities
for learners. Brannon and Essex (2001, p. 36) state that “asynchronous
communication can be helpful for encouraging in-depth, more thought-
ful discussions, communicating with temporally diverse students, holding
ongoing discussions where archiving is required and allowing all students
to respond to a topic.” As well as the advantages, it also has some draw-
backs such as lack of immediate feedback, not checking often enough,
students feeling a sense of social disconnection (Brannon and Essex 2001).
One prominent study conducted by Brannon and Essex (2001) com-
pared the two models of collaborative learning: asynchronous communi-
cation and face-to-face communication. Computer conferencing system
and text-­based discussion were used as a medium of asynchronous com-
munication. The results revealed that “a technology-supported collab-
orative environment is an effective means of learning and conducting
complex group work. However, it also shows us that people prefer to
interact in a face to face manner due to the slow process of social change”
(Brannon and Essex 2001).
To sum up, it is evident that asynchronous communication in the con-
text of computer-mediated communication provides opportunities for
collaboration and that language learning has gone beyond face-to-face
 9  Presenting and Writing Up Action Research 
   227

interaction. Moreover, computer-mediated communication has been mea-

sured as a facilitator of interactive communication for language learning,
therefore making it all the more important for EFL settings such as ours,
where a communicative approach to English language teaching is adopted.

• Research questions: When a gap has been identified, this forms the
basis of the research question or questions. The research question/s
should be clearly set out, and should be addressed point by point. It
might be necessary to explicitly show the relationship between the
purpose and research questions.
• Setting: As much detail as possible should be given about where the
research is taking place—country, type of institution, and so on. Since
AR is a contextualized form of research, results may not be generalizable
to a wider context. Readers will read a study with the contextual condi-
tions and constraints in mind. The knowledge about the context will
help them understand the research in a more meaningful way. For exam-
ple, they will evaluate the results better if they know where the research
took place. (Remember Chap. 4 for the discussion of context.)
• Participants: Also give as much detail as possible about the participants
in the study—ages, genders, nationalities, and so on. Since these peo-
ple are the source of information, it is necessary to show that partici-
pants who can provide the best information for the research have been
chosen. If it is a class, it is necessary to justify why that class was cho-
sen, not another.
• Data collection procedures: Here the research method/s should be speci-
fied—whether mainly quantitative or qualitative, cross-sectional or lon-
gitudinal, experimental or quasi-experimental collection procedures,
and so on (see Chap. 3 for more details of research method types). This
section is important in that it is this part that shows how reliable and
valid the research is. The readers will judge the quality of the research by
considering the methodological procedures followed. Many papers sub-
mitted for publication are rejected because they do not report well
enough the ways in which the research was carried out. How the data are
going to be collected should be specified, for example, in class, online,
when, how often, and so on. Data collection instruments should also be
specified, for example, a questionnaire, observation schedule, journals,
or other materials. If it is to be ­experimental or quasi-experimental, how
228  Developing Language Teacher Autonomy through Action Research

will the participants be grouped? Will assignment be totally random, or

will intact classes or volunteers be used? Who is/are the teacher/s? If
there are to be more than one teacher, how will the confounding variable
of teacher individuality be controlled for? All of these details are impor-
tant and may influence the results. Sometimes, if this section is rather
big, it may be split into more than one section, with headings such as
“Research methodology”, “Instrumentation”, and so on. Sometimes,
setting out data collection tools as a table (as per Table 9.1) may help to
clarify exactly how data collection was done:
• Ethical issues: Here any ethical issues to do with data collection should
be specified, such as obtaining approval from the institution, assuring
confidentiality, providing information, voluntary participation,
obtaining consent, storage of data, and so on. (See Chap. 5 for more
details on this important issue.) Many journals will not accept articles
where these conditions are not met. It is better to attend to these issues
before the study is started: it is too late afterwards.
• Data analysis procedures: In this section, data analysis procedures
should be specified, for example, means, medians, differences, corre-
lations, and so on. The analysis tools will vary according to the type
of data, but, as a rule of thumb, parametric statistical tests (such as
means, Pearson’s correlation, t-tests, ANalysis Of VAriance
(ANOVAs), etc.) can be used for numerical data (such as ages, exam
scores, etc.) as long as they are normally distributed. Nonparametric
tests (such as medians, Spearman’s correlation, Mann–Whitney U
test of difference, etc.) should be used for non-numerical data (such
as Likert-type questionnaires) or for numerical data which is not nor-
mally distributed. If a grounded approach is being used, the coding
procedures should be specified, and any procedures used to analyse
qualitative data (such as case studies or ethnographies) should be

Table 9.1  Data collection tools

Quantitative data collection tools Qualitative data collection tools
Close-ended questionnaires Interviews
Test scores Observations
Surveys Document analysis
Dairies
Lesson recordings
 9  Presenting and Writing Up Action Research 
   229

noted. What is said here should match what is in the results, and the
results should match what is specified here.
• Results: Here the findings should be reported. As a general rule, com-
ments or opinions should not be given at this point. This comes in the
“Discussion”, although sometimes these two sections are put together,
especially in qualitative studies. Data analysis procedures should be
described (see Chap. 6). The results should be checked against the
research questions to make sure everything has been included.
• Discussion: This is where the results are discussed and commented on,
and the researcher can give his/her own thoughts and opinions, com-
pare with previous research findings, and so on. In this section, it is
possible to discuss what these findings might mean, the benefits, and
what was learnt from engagement in this research. This section requires
reflection upon the analysed data. This is actually the reason the study
was conducted. Now readers can be told how or whether the purpose
of the research has been addressed, and about new understandings that
could inform teaching practices. This part can follow such strategies as:

• Comparing and contrasting the understanding of the research focus

before, during, and after the research
• Comparing and contrasting findings with those in existing literature
• Giving specific examples of how things might have changed or can
change in the future
• Discussing how research experience will contribute to teaching
• Discussing what new perspectives the research has opened for
reflection
• Introducing feelings and opinions about learning through doing
research
• Making an overall reflection on the research experience

• Implications: Here the meaning of the findings for classroom practice

or the various stakeholders who may have an interest in the results
(such as parents, educational authorities, etc.) is discussed.
• Limitations/suggestions for further research: Every study has some limi-
tation or other, perhaps limited numbers, context, research methodology,
or whatever. In order to avoid the negative connotations of “limitations”,
230  Developing Language Teacher Autonomy through Action Research

the section can be called “Suggestions for further research” or some such
phrase. Here ideas for extending the study or looking at related areas or
questions which arose during the study might be suggested. This can be
a very useful section for those looking for a research topic.
• Conclusion: The conclusion should summarize the article and empha-
size the main points. This is an extremely important section. When
readers are skimming for suitable articles to cite for their own research,
they will typically go for the title, the abstract, and the conclusion. If
they prove to be interesting, they will read the rest. There should be no
new references in the conclusion.
• References: The American Psychological Association (APA) style is
most commonly used in the Applied Linguistics field, although this is
not the only possible model. The important thing is that, whatever
style is used, should be consistent. If submission to a journal is planned,
the author guidelines should be checked for the style appropriate to
that particular journal. A brief summary of the main APA referencing
types can be found at the end of this chapter.
• Appendix/ces: Here, everything (e.g. questionnaires, checklists, mate-
rials) that might make the study replicable, or which provides informa-
tion that is not necessary in the main body of the article (e.g. statistical
or other details that not every reader might want to know about or
understand), should be included. Some journals make this kind of
material available online, especially if it is bulky and might make the
article over-length.

These points are summarized in Table 9.2.

Less Formal Ways of Writing Up Research

Informal ways of organizing articles could be more reader-friendly, con-
sidering that teachers and many novice researchers are intimidated by
engagement in research, and have no real interest in publishing. The
following outline of paper organization could sound less daunting with
less jargon than academic research conventions. The guideline in Table
9.3 was developed for Dikilitaş et al. (2016) for an edited collection of
teacher research studies.
 9  Presenting and Writing Up Action Research 
   231

Table 9.2  Major sections of a research article

Abstract: Summarize the research with key characteristics such as purpose,
data collection tools and analysis, and key findings. Journals usually have a
word limit, so keep to that
Keywords: Select keywords which make it easier for the article to be found
online by a researcher looking for references on the same topic
Introduction: Provide an overall design of research and synthesize key
discussions and the significance of the research among all related ones.
Identify general and specific purposes of the research too, and outline what
the article intends to do
Literature review: Write a literature review which recognizes other published
sources. This section can be organized around major themes or
chronologically. The literature needs to be relevant to the points being
made in the study, and should identify the gap/s in existing research
Research questions: The research questions should address the gap/s
Context: In the interests of establishing credibility, the context section should
give as much detail as possible about the setting and the participants
Data collection: This section should be specific about how the data were
collected, the methods and the instruments that were used, and should
include how ethical issues were dealt with
Data analysis: Here the types of procedures used to analyse the data (e.g.
means, medians, correlations, differences) should be specified and justified
Findings/results: The findings for each research question should be presented
here
Discussion: Discuss the findings by synthesizing them with the previously
reported findings in the literature on similar topics. In this way, the place
and significance of the findings can be located among all other relevant
ones
Implications: The meaning of the findings should be discussed here, whether
on a theoretical or on a practical level. A well-organized academic research
study can propose theoretical implications as well as practical ones, which
might inform professional researchers and educational practitioners
Limitations/recommendations: Any limitations of the research (every study
has some, e.g. limited numbers, limited context) may be acknowledged here;
areas for ongoing research can be suggested
Conclusion: This section is for final remarks which summarize the study and
include reflection over what has been found
References: List the references cited in the text alphabetically, following the
style specified by the journal
Appendix/ces: Include any extra material relevant to the topic of the study,
especially any materials (e.g. questionnaires, interview schedules) which
might help another researcher to replicate the study
232  Developing Language Teacher Autonomy through Action Research

Table 9.3  Informal research organization

Main focus:
 Introduce what will be discussed in this article, highlighting the purpose
clearly and saying why this topic was chosen for investigation
Background:
 Discuss the major issues underlying the research. This section could consider
contextual issues and include a brief review of relevant literature
Research methodology:
 If primary data has been collected, describe the research methodology,
including information about research participants, research approach and
methods, data collection tools, analytical procedures, and any ethical issues
Findings and discussion:
 Present findings and discuss them reflectively, addressing the research
questions and perhaps relating findings to the literature
Reflections and looking into the future:
 Reflect on how this research has influenced teaching practices and discuss
how this research will help continued development as action-researcher

Benefits of Publication
Although not everyone may wish to pursue publication, the benefits of
publishing results for wider distribution are summarized in Table 9.4

A Personal View: Carol Griffiths

Although it might not be everyone’s “cup of tea”, and it is far from the only
option, many researchers do in fact have the ultimate ambition of seeing
their work in print. For them, the ultimate purpose of undertaking the
hard slog of conducting research, in the face of many other more “fun” and
relaxing ways of spending the time, is to get published in a journal. A word
of caution for these hardy individuals, however: it is not easy! A few words
of advice, therefore, from one who has been through the press, as it were.
If you are serious, you need to know the market. If you want to find some-
one to publish your research, you need to do some serious reading. Most
journals are available online these days, and you are looking for a journal
which publishes your kind of topic. Don’t forget to include some of your
target journal’s articles, especially the most recent ones, in your citations
(this contributes to their impact factor). Read the author guidelines and
 9  Presenting and Writing Up Action Research 
   233

Table 9.4  Benefits of publishing action research, summarized from Burns (2014)
Crookes (1993) Other teachers can make use of the insights gained
A body of knowledge that will help teachers
contextualize their own work
Burton and Mickan Leads to teachers’ feeling “more at ease with reading
(1993) more widely”
Encourages an interest in professional reading
Freeman (1996) Becoming part of a wider, more open research community
Contributing to “discussions of policy and disciplinary
knowledge”
McDonough and Findings “can be brought to a wider audience and can be
McDonough subjected to critical analysis by their peers”
(1997)
Freeman (1998) The potential for other teachers to learn from research
Edge (2001) Teachers “should communicate their experiences to
others” to “spread first-hand accounts of …significant
outcomes”
Bailey, Curtis and Academic researchers as well as teachers “benefit from
Nunan (2001) listening to other teachers’ stories”
Richards and “Part of the philosophy of action research is sharing the
Farrell (2005) findings with other colleagues” in order to lead to better
understanding
Allwright and The potential to inform decision-making and lead to wider
Hanks (2009) theoretical insight in educational contexts
Findings may “encourage others to joining the debate
and in the search for yet deeper meanings”
Barkhuizen (2009) Benefits for “curriculum developers, school-policy makers,
and the wider language teaching community”
An opportunity to contribute to current issues in the
field of language education
Borg (2013) Dissemination of a “basic characteristic of all research” in
“the many varied formats, oral and written, formal and
less formal, through which they can make their work
available for public scrutiny”
Burns (2014) Publishing results shows that they have value, makes
knowledge available for public scrutiny, and become
sources of inspiration or support for further research

follow them to the letter, since authors who can’t be bothered doing this
are sending warning signals that they are bad risks when it comes to the
kind of tedious detail that will be required further down the line!
And what about rejection? This needs to be included because this aspect
is often neglected or glossed over. Remember that the “top” journals only
234  Developing Language Teacher Autonomy through Action Research

accept about 10% of the articles submitted to them. This means that if
you submit to these journals, your chances of rejection are 90%. It would
be like sitting an exam knowing you only have a 10% chance of passing.
Does this mean you should not try? Of course not!
But you need to accept realistically from the beginning that your
chances of acceptance are not high. Remember also that the first time
you hear back from the editor, it will almost certainly be a rejection. This
is because one of the factors used to calculate a journal’s impact factor is
its rejection rate. When you get the first rejection, it is important to view
the sub-text. If they say something like “Not suitable for this journal”
and, perhaps suggest another journal, it means they are not interested. If
they invite you to revise and re-submit, it means they are interested, so it
may be worth following their suggestions and trying again.
Let me give you two examples from my own experience. Late in 2015
I had an article on learning strategies (my specialty) published in a top
SSCI journal (ELTJ). When I first submitted this article to a different
journal, they returned it saying it was not theoretical enough, and also
it was not a topic that they were interested in pursuing right now, but
thanks for thinking of them and they hoped it would not discourage
me from submitting in the future, etc., etc. In other words, it sounded
pretty final. So I revised it according to the requirements of ELTJ, but
they returned it saying it was not practical enough! However, they sug-
gested that I might re-focus it emphasizing the pedagogical implications.
So I did that. In fact I did it twice more before it was finally accepted
(Griffiths 2015).
Another example is my 2008 book, Lessons from Good Language
Learners (CUP). When I first approached CUP, the editor’s reaction was
that she thought what I had in mind didn’t sound academic enough for
CUP. I continued to work on it, and about the time I had a complete
draft ready another more pedagogically focussed publisher was advertis-
ing for publications, so I submitted a draft there. I received back the worst
review I have ever had for anything either before or since. The reviewer
was absolutely scathing, describing his/her “agony” at having to read such
“academic clap-­trap”, and why would anybody submit such “rubbish”
to this particular publisher. It was, quite frankly, an extremely unprofes-
sional review, and if I received such a review today, with confidence born
of more successful experience, I would complain about it in terms at least
 9  Presenting and Writing Up Action Research 
   235

as strong as those used. At the time, however, it depressed and demoti-

vated me almost completely, and I remember sitting at my desk with the
manuscript on one corner and the rubbish bin just under the corner,
and came close to consigning it to the trash. However, it was Friday, so
I decided to think about it over the weekend, and closed my office door
with the manuscript still sitting on the corner of the desk above the rub-
bish tin. Over the weekend I got to thinking that if it was too academic
for a practitioner-focused publisher, maybe it would be academic enough
for CUP, so, on Monday I contacted the CUP editor again. This time
she was interested, so I sent her the draft. After several revisions and re-­
revisions, it was sent out for review, and this time I received one of the
very best reviews I have ever received—almost everything was good about
it, so CUP published it, and it has received good reviews almost every
time it has been reviewed ever since.
I tell these personal narratives in the hope that it might encourage
those who want to publish to persevere. Even for experienced authors, it
is not easy, so, if you really want to do it, do not be too quick to give up.

Questions for Consideration

1. Do any of the conferences noted here appeal to you as a venue for pre-
senting your study? What are the reasons for your preference?
2. What do you see as the advantages/disadvantages of presenting research
to an audience?
3. Which of the presentation modes noted here (or others you might think
of) might best suit your own needs or preferences?
4. What do you see as the comparative advantages or disadvantages of
writing a book, a chapter, or an article?
5. Which of the sections of an article do you think is most challenging?
6. Which of the benefits noted in Table 9.4 do you find personally most/
least compelling?
7. Do you have any personal experiences of presenting or publishing that
you would like to share?

Task
Decide how you wish to present your research (e.g. as an in-house talk, an
article, poster, PowerPoint presentation) and prepare the materials you
need to do this.
236  Developing Language Teacher Autonomy through Action Research

APA Style
APA stands for the American Psychological Association. It is a very
common style in social sciences (though be aware that it has changed over
the years), and it consists of two main types:
In text
Studies have found that strategies are related to progress in language
learning (Griffiths, 2015)
According to Soruç and Griffiths (2015), uptake of spoken grammar
features is related to identity
Good learners use strategies (Cohen, 2011; Griffiths, 2013; Oxford,
2011; Rubin, 1975)
For direct quotes, page numbers are required:
“Teachers should remember that strategy development is not linear
but spiral” (Griffiths, 2013, p. 176)
“There was also a high degree of accord between the teachers’ impor-
tance ratings and the strategies the students reported using highly fre-
quently” (Griffiths, 2013, pp. 176–177)
Reference list
Book:
Griffiths, C. (2013). The Strategy Factor in Successful Language Learning.
Bristol, UK: Multilingual Matters.
Journal article:
Griffiths, C. & Cansiz, G. (2015). Language learning strategies: an
holistic view. Studies in Second Language Learning and Teaching, 5(3),
475–495
Online article:
Griffiths, C. & Alymidin, J, (2015). Using narratives to prepare an
integrated skills lesson in ELT: Black June. www.hltmag.co.uk
Chapter in book:
Griffiths, C. (2012). Learning Styles: Traversing the Quagmire. In
S. Mercer, S. Ryan & M. Williams (Eds), Language Learning Psychology:
Research, Theory & Pedagogy (pp.  151–168). London, UK: Palgrave
Macmillan.
Other variations on these basic types can usually be found online.
 10
Researcher Narratives

This chapter will include a number of narratives which, it is hoped,

will comfort or inspire others with the knowledge that you are not
alone in whatever you may be feeling: others have also been down the
same road.
The narratives in this chapter are provided not for replication, but for
developing insights into the process of doing research. The narratives
include the authors’ internal feelings while they are doing it and their
comments on the research as well as the benefits they gained through
doing it. There are three types of narratives: One is the group of narra-
tives on the actual studies whether published or not. Another consists of
negative and positive personal narratives of teacher researchers. The last
group comprises meta-narratives in which the authors discuss retrospec-
tively studies conducted among teachers or by others and include their
comments.

© The Author(s) 2017 237

K. Dikilitaş, C. Griffiths, Developing Language Teacher Autonomy
through Action Research, DOI 10.1007/978-3-319-50739-2_10
238  Developing Language Teacher Autonomy through Action Research

Actual Studies
 hat Are the Most Important Strategies for Language
W
Skills Development?

Carol Griffiths
In the course of a long teaching career, I have found myself teaching
many different subjects to many different types of students. One such epi-
sode that I remember quite clearly was teaching a research methods course
to a class of ESL (English as a second language) students in Auckland, New
Zealand. The class included students from many different nationalities, but
the majority were either Chinese or Indian in origin, and for almost all of
them, English was not their first language (L1). Although in theory the
students were at a level where their English was sufficient for them to cope
with the demands of quite a high-level paper with a great deal of quite dif-
ficult vocabulary, in reality, many of them were clearly struggling.
As a result of this observation, I got to wondering how they might be
helped, and I formulated several key questions:

• What strategies did they already use?

• Was there was any relationship between certain strategies and success-
ful outcomes for the course?
• Which strategies might they usefully be encouraged to employ in order
to help them to understand the content of this quite demanding paper?

In order to explore these questions, I constructed a questionnaire

which I called the “Language Skills Development (LSD) Questionnaire”,
consisting of ten strategy statements about each of the four skills (read-
ing, writing, listening, speaking), which students were asked to grade
from 1 = very low to 5 = very high according to the frequency with which
they used each one.
In fact, in addition to the research questions, I had an unspoken
hypothesis: in my experience, the best students I have ever known read
in English, and although many students are quite resistant to reading, I
was quite convinced, based on previous students I had known, that the
amount of reading they did was a factor in their success or otherwise.
 10  Researcher Narratives 
   239

When the questionnaires were completed and collected and the data
analysed, to my great satisfaction, I did in fact find that the only skill
group which correlated positively and significantly with course results
was the reading strategy group, accounting for around 18% of the varia-
tion in final scores. Of the reading strategy group, the item by far the
most strongly correlated with successful outcomes was reading strategy 3
(I find reading material at my level), accounting for around 35% of the
variance in results.
Out of the 32 students who completed the questionnaire, there were
6 who achieved an A pass. When these students were interviewed, all six
mentioned using reading as a key strategy for developing their language
skills in English. Reasons given for considering reading a useful strategy
included:

• Reading is a source of new vocabulary

• Reading provides examples of grammar
• Reading provides examples of how language is used
• The reader has more control than a listener
• What is learnt when reading can aid comprehension when listening
• Reading provides a model of correct language which can then be
employed when writing or speaking

Types of reading these students considered useful included:

• Newspapers
• Magazines
• Textbooks
• Novels
• Graded readers

I therefore felt a sense of personal satisfaction with the results of this

study: they confirmed something I had long believed about the value of
reading when learning language. Of course, by the time the question-
naires had been collected and analysed and correlated with the final
scores, the findings were not helpful for this particular class of students:
the course was finished.
240  Developing Language Teacher Autonomy through Action Research

Have I used the findings to actively promote the value of reading in

the years since? I have to admit that, amidst the hurly-burly of life as a
teacher, where preparing and delivering lessons and assessing results has
to be the priority, I have not been as proactive in this regard as I might
have been. But just writing this narrative has brought back to my mind
the effectiveness of reading in the target language, and I find myself mak-
ing a new resolution to take more initiative to promote what I am con-
vinced is one of the most powerful strategies available.
But maybe such resolutions are easy, writing this as I am in my sum-
mer break, removed from the daily realities of a real-life classroom. Let’s
see if I can maintain my motivation to carry out this resolution once
classroom realities reassert themselves.

 hy Do My Students Have Difficulty

W
in Comprehending Authentic Texts?

Kenan Dikilitaş
Bahçeşehir University, İstanbul, Turkey
I have a long teaching career at several stages of education ranging from
kindergarten to graduate levels, and I have had a number of puzzles and
questions when working with so many different teachers and students. One
of the issues which puzzled me a lot was the reason that Turkish learners
have so much difficulty in comprehending authentic academic texts at an
upper-intermediate level. The students often have a high level of English,
particularly in tests requiring grammar and vocabulary in the form of mul-
tiple choice tests. However, when it comes to reading and comprehending
authentic texts about academic topics, most display difficulties.
In order to address the issue, I developed the following question which
I thought would be helpful for investigating and understanding the issue.

1. What do my students think are the major causes of inability to read

and understand authentic academic texts?

To be able to answer this question, I planned some strategies, which

were integrated into the classroom practices. At first, I initiated a whole
class discussion and took note of what each student was saying on the
 10  Researcher Narratives 
   241

board about the causes of difficulties in reading comprehension of

authentic texts. In almost 20 minutes I wrote 37 short sentences that
20 students reported verbally. After the completion of the discussion, I
asked the students to categorize the sentences under the main themes and
to indicate how many times each was reported. The emerging categories
developed by the students included:

• Unknown vocabulary (9)

• Long and complex sentences (7)
• Disengaging topics (7)
• Low motivation for reading (6)
• Little interaction while reading (8)

I then asked them to further categorize these five themes if they could.
I told them to create major titles. They discussed in groups and reported
the following:
Categories Themes
Unknown vocabulary Vocabulary
Long and complex sentences Grammar
Disengaging topics Interest
Low motivation for reading Motivation
Little interaction while reading Interaction

The major aim of this further categorization of themes activity was to

make them discover the actual causes and raise their awareness, which
is one of the main skills in comprehending academic texts. In addition,
labelling ideas could be a good strategy for overall comprehension of
texts, which can facilitate understanding. I found that they enjoyed such
discussion and negotiation.
Having observed and learned from the students that the activity was
engaging and useful, on the following day I decided to make use of this
outcome by integrating the same activity in the reading text. I brought
three texts and asked them to choose one that they found interesting. I
did so because I wanted to address the “disengaging topics” problem that
they reported. They chose one about Technology. The text included eight
different paragraphs of 70–80 words each, focusing on different aspects
of the same topic.
242  Developing Language Teacher Autonomy through Action Research

I designed an activity on the basis of the causes they reported in the

discussion and categorization activity. I addressed each of the concerns
raised in the following way:
Categories Instruction
Unknown vocabulary (9) Do not try to understand each word you
don’t know
Long and complex sentences (7) Do not try to understand the meaning of
each sentence separately
Disengaging topics (7) Select one of the topics
Low motivation for reading (6) Select the topic you want to learn more
about
Little interaction while reading (8) Individual followed by pair work and
group work activities

After the text was chosen, I asked them to read the text and decided to
discuss only one paragraph. After all students had read their paragraphs, I
initiated a “union-ring”, which involves forming embedded rings, where
students can revolve face-to-face. Ten students were inside the ring and
the other ten outside. I gave one minute for two students (half for each)
to tell each other about the paragraph they read. This gave them more
opportunities for interaction. So each student talked about his/her own
paragraph ten times and listened to their friends presenting each para-
graph to them as well.
After the union-ring activity, I formed five groups of students who
would thematize each paragraph and make a list in order to create an
outline of the specific topics of each paragraph. At the end of the lesson
we discussed the influence of such an activity on their reading. The issues
they highlighted were:

• It was empowering to discover the causes though an inductive activity

where they worked with the data they generated
• They were happy to be granted an opportunity to choose the topic
they were going to read
• They found it enjoyable to read and comprehend the text coopera-
tively which helped them interact and learn from one another while
also helping others understand
 10  Researcher Narratives 
   243

• The lesson created a dynamic environment where they could display

active participation and take control of the lesson with little teacher
intervention
• They developed reading strategies especially about unknown vocabu-
lary, whose meaning they either discovered through negotiation or
learnt from others
• They reported developing confidence in reading through collaborative
activities in the classroom

It was an invaluable research experience for me to engage in learners’

challenges and develop classroom-oriented and learner- and learning-­
centred research designs. I helped students identify their challenges and
realize that they could overcome them through autonomy-developing
activities in which they took part actively. I also realized that I could
minimize my instructional time for reading texts and maximize learn-
ers’ engagement in activities through different interactional patterns that
could create learning though discovery.

Learning Strategies in China

Ying Tang(Rose)
Shenzhen Xin’an Middle School, China
As an English teacher, like other teachers in mainland China, I worked
overtime for decades, busy with giving lessons and after-class coaching
with the hope of helping students receive high scores in all kinds of exam-
inations. Although I made some achievements, I saw that taking every
opportunity to give extra lessons resulted in students being tired from
heavy homework, which led to their feeling sleepy in class, even tired of
going to school and giving up learning. Both learners and teachers were
struggling to learn and teach by painstaking efforts and hard work.
In mainland China, it is a common phenomenon that many learners
are unable to communicate with foreigners after having learned English
from primary school to high school and even to college for more than
a decade. People attribute students’ failure in learning English to lazi-
ness, lack of aptitude for language learning, and teachers’ poor teaching.
244  Developing Language Teacher Autonomy through Action Research

Although many Chinese students have worked very hard at English by

rote learning for many years, their English remains very poor. Seeing this,
I was deep in thought.
Research has shown that learning strategies play a vital role in language
learning, but learners and teachers at my school did not know about lan-
guage learning strategies, and no strategy training was carried out. I got
to wondering how middle school students in China might be helped, and
I formulated several key questions:

1 . What are effective English learning strategies for Chinese students?

2. Which strategies might students employ in order to help them to
enjoy learning English and having lessons?

In order to investigate the effect of strategy training on achieve-

ment in English, two senior classes were chosen: Class 10 (N = 45) was
the control, and Class 9 (N = 50) was the experimental group which
received instruction in 30 different strategies. At the end of the previ-
ous term, the average score for Class 10 was 73, whereas the average
score for Class 9 was 76. After the training, the average for Class 10
was 77.4 (an increase of 4.4), while the average for Class 9 was 85.3
(an increase of 9.3). In other words, Class 9, which received the treat-
ment, improved considerably more than the control group (Class 10).
In addition, 89.1% of the students said they liked the training, 87.2%
thought it had given them more confidence, 85.5% agreed it had made
learning easier and more interesting, and 90.9% thought it had helped
improve their English.
To my great satisfaction, those who received the training did much
better in their exams than those who did not receive the training. This
suggests that learners can gain effective English learning strategies by
means of classroom instruction, which helps them improve their English
and also enjoy learning. I therefore felt a sense of career satisfaction with
the results of this study. It has helped me realize my dream of doing
something and making contributions to English teaching and learning in
China. I have decided to continue with my field research and do my best
to make more contributions to English teaching in China and actively
promote our strategy training.
 10  Researcher Narratives 
   245

 poken Grammar: An Approach to Understanding

S
Classroom Interaction?

Simon Mumford
University of Economics, Izmir, Turkey
This is a short account of how I, as a teacher, came to new understand-
ing of the role of a specific aspect of language, namely, spoken grammar.
This involved me initially taking a position in favour of teaching spoken
grammar. However, based on interaction with and observation of my
students, I abandoned this position, instead seeing this new language
knowledge as an approach to understanding classroom interaction, rather
than a content to be taught directly.
The story started a decade ago, when I was fortunate enough to be
invited to work on a book proposal for teaching spoken grammar, which,
at the time, was considered to be an important development in English
Language Teaching. These were exciting times; 2006 was the year of the
publication of the definitive Cambridge Grammar of English (CGE), writ-
ten by Carter and McCarthy. This corpus-based grammar contained the
most comprehensive account so far of the features of spoken English.
Expressions such as headers, tails, co-construction of communication,
and ellipsis were identified as integral features of the language. My two
co-authors and I were aiming to write a book of classroom activities
that would allow students to practise and produce these newly revealed
features.
However, not everyone was so enthusiastic. Some researchers pointed
to the difficulties of consciously teaching ephemeral features such as hesi-
tation or flexible word order, arguing that awareness raising was the best
that could be hoped for. Of greater concern was the argument by English
as Lingua Franca (ELF) scholars that such language was too closely
­associated with the native-speaker variety, and therefore irrelevant to the
non-­native majority of English speakers.
This led me to consider the reasons for actively teaching spoken gram-
mar, and to make the case for this point of view. I realized that some fea-
tures were more useful to learners than others. Useful features included
ellipsis, that is, leaving out unnecessary words (you going out tonight?)
and using simple phrases, (a new house, six bedrooms instead of a new six
246  Developing Language Teacher Autonomy through Action Research

bedroom house). These two features can increase fluency by reducing the
number of words needed and lightening the cognitive load, respectively,
in theory, leading to greater economy and efficiency in language produc-
tion. In contrast, other features, such as being able to use native speaker
exclamations (Really?, Oh dear!), seemed less appropriate for teaching.
The classification between the useful and less useful features evolved into
an argument for the teaching of such forms, eventually appearing as an
article in the ELT Journal in 2008 entitled Spoken grammar: The case for
production.
However, there now remained the difficulties of putting theory into
practice, which, in my initial enthusiasm, I had not really considered.
Where to start? There was no time in my syllabus for teaching extra con-
tent, and no clear guidelines on how to teach it, even if the time were
available. So, at that point, with the rejection of our book proposal add-
ing to my frustration, I abandoned all efforts to actively teach spoken
grammar. Nevertheless, I increasingly realized that spoken grammar was
intrinsically part of the language I was teaching, and, therefore, far from
irrelevant. Teaching involves recasts, repetition, hesitation, ellipsis, co-­
constructed dialogues. My own hesitations, false starts, and repetitions
(Er...turn to...er…open open your books at page 8) and non-standard word
order (Exercise one we are doing next) were inevitable features of every
lesson.
Furthermore, rather to my surprise, discussions with students revealed
that they were in fact generally aware of these aspects in their own lan-
guage. Many of the features of spoken grammar, such as flexible word
order, hesitations, and repetition of words, are universal, not just related
to English. However, students noted that they were not specifically taught
these features in their own language, Turkish; they understood that these
features arose from the pressures of real-time communication. So, even if
it were possible, direct teaching of spoken grammar in English would not
necessarily be welcomed by these students, who were focused on learning
vocabulary and acquiring the necessary writing, reading, and listening
skills to pass exams.
Through these discussions, and reflection on spoken grammar, I real-
ized that although not suitable for teaching, the use of spoken grammar
was inevitable in the classroom, and I decided to investigate, in a very
informal way, whether features of spoken grammar arose naturally in the
 10  Researcher Narratives 
   247

classroom. I teach EAP (English for Academic Purposes) in an English

university in Turkey. Students interact with the teacher and each other in
the medium of English, so I set out to observe teacher–student interac-
tions, and also student–student interactions. I did this through noting
interactions in real time, and also watching recordings of lessons. When
considering teacher–student interactions, here are several points I noticed:

• A student began to reply to a teacher’s question, then hesitated and

stopped. But this is natural; people do not speak in sentences. Maybe
he considered that he’d said enough for the teacher to understand that
he knew the answer.
• In class, a student began to answer a question, but a second student
took over. A teacher might consider this as a rude interruption, but
passing turns in mid-utterance is quite natural in everyday speech.
Perhaps the first student will take his turn back later, or the turn will
pass to another student.
• A student meets a teacher’s question with silence. In an informal con-
text, a filler (er…), or gesture (shrug), would suffice to pass a turn.
Why assume the student has failed to convey this message because of
the absence of words?

In these situations, classroom behaviour that I would previously have

associated with a lack of ability to communicate, or a lack of cooperation,
or even rudeness, could be interpreted as being entirely consistent with
the norms of spoken grammar.
With regard to student–student interaction in pair work, I hypothesized
the following: When students were unmonitored they would use a more
informal spoken grammar, that is, co-constructed dialogue, flexible word
order, devices such as laughter and gestures, while when being monitored,
the focus would shift from fluency and communication to accuracy and
display language. The problem here was the observers’ paradox, in that
it was impossible to capture unmonitored student interaction, because
when they were aware of being monitored, they would be likely to switch.
However, an informal discussion confirmed that some students agreed
that being monitored changed the nature of the interaction, resulting in a
greater focus on the code rather than the message. This was not necessarily
seen as a bad thing by students, as a switch to a more conscious attention
248  Developing Language Teacher Autonomy through Action Research

to accuracy focus may result in a different quality of language development.

This switch was largely seen as inevitable. Teachers are expected to monitor
student performance; this is their accepted role.

Implications

This was not a formal research project, and there was no attempt to
quantify students’ attitudes to spoken grammar. However, this narrative
reveals a process of discovery, starting from an attitude based on a par-
ticular interpretation of language, which became adapted in the face of
reality. Interaction with students confirmed that spoken grammar would
probably not be suitable content for teaching in this context, an EAP
course in an English medium university, with students focused on the
goal of passing (written) exams. However, knowledge of spoken gram-
mar allowed for a reinterpretation of student behaviour, and provides
a framework for understanding the difference between monitored and
unmonitored speaking pair work, in which students may switch from a
more conversation-like discourse to a more formal, and arguably, artificial
one. It would be interesting to explore these areas further, to understand
how the teacher’s knowledge of spoken grammar could affect classroom
interaction, and explore further the different qualities of monitored and
unmonitored student interaction.

Teachers’ Personal Narratives

Discovering Your True Potential

Koray Haki Akyazı
Like a caterpillar, who during the chrysalis stage has the potential to
turn into a butterfly, action research is a way for teachers to liberate them-
selves from their cocoons.
After gaining a Certificate in English Language Teaching to Adults
(CELTA) in 2012, I started working at a private university, where I was
encouraged to participate in the teacher research programme. Initially, I
remember being rather critical of such an approach to teacher development
 10  Researcher Narratives 
   249

as I had just completed an intensive craft model training course. Until

then, I had no conception of what action research was, and was expect-
ing more CELTA-style practical workshops in my growth as a language
teacher. I couldn’t quite see the impact of action research on my pedagogic
practice. It took a year and a lot of self-questioning and reflection to come
to the conclusion that there had to be more to it than this. I had only seen
the tip of the iceberg during my initial exposure that first year, looking at
the best way to teach phrasal verbs through controlled groups and pre-/
post-tests. The following year I asked myself what had I gained from this
experience, and so decided to shift the focus of the research to myself. My
second venture into action research was the turning point in my develop-
ment as a teacher researcher. I invited a colleague to work with me in order
to observe each other’s lessons and we noticed differences in the way we
approach grammar instruction; he used a more deductive approach and I
an inductive approach. In later lessons, I experimented with explicit gram-
mar instruction to various extents till my learners and I felt happy with the
way lessons were done. We observed each other’s lessons three times and
took specific notes of how grammar was taught. After the observations, we
discussed grammar teaching practices in the presence of the teacher trainer
in the school. In addition to the observations, we also interviewed with
the students about their views of specific grammar teaching approaches.
This classroom observation practice was coupled with much reading,
until I gradually realized the true essence of action research. Finally, action
research as a professional development tool began to make sense.
In the following years, I gradually became more aware of how action
research differed from academic research. Knowledge generated as a
result of the research was not a prime outcome, but it was the process
of development which was more important to me, and to my learners. I
was later introduced to the notion of exploratory practice (EP) as a form
of teacher research. With a more ethical focus, I became more aware of
the learners in the process of researching together with the class, rather
than researching them as subjects. In my previous action research, I con-
sulted my learners on how I could go about the initial idea of introducing
authenticity into the class, and they were very forthcoming with their
own suggestions. They really appreciated being listened to, and, in fact,
I invited two students to join me as co-presenters in the poster presenta-
tion of the paper, thus continuing with the theme of authenticity.
250  Developing Language Teacher Autonomy through Action Research

I also noticed an increase in my self-efficacy in research skills, and I

had been doing a lot of reading over the years, so much so that I thought
I might as well be studying towards a master’s degree. When the oppor-
tunity arrived, although my BA was in Business Administration, I was
able to demonstrate my ability to study a master’s in English language
teaching. In fact, my thesis was on how other language teachers perceived
teacher research, and how engagement in teacher research affected their
teaching. I found that the teachers with prolonged engagement in teacher
research were better able to benefit in terms of both pedagogic knowledge
and pedagogic practice.
I would recommend teachers to engage in research no matter where
they are in their career. With support from critical friends or a mentor,
you can become the teacher you have always dreamt about being. It may
not be easy at times, but it is well worth it in the long run.

 xploratory Practice as Tool for Teacher/Learner

E
Education and for Discovering Inner Learner/
Teacher Self

Kerim Biçer
My journey as a teacher researcher had begun long before I was for-
mally introduced to EP when I did my first ever action research on learner
(de)motivation. Back then, I did not possess half the courage, valoriza-
tion capability, and insight I do now yet it was still such an empower-
ing experience to have been the investigator in my classroom striving to
find answers to my own pedagogic/instructional problems that would
constantly arise. Nonetheless, it felt as though there was still something
missing—a missing link I always kept looking for.
I first came across EP by sheer coincidence during a local British
Council teacher-training workshop I was invited to attend. At first, I
had serious reservations as to whether it should be a similar unprepos-
sessing experience (of detachment and mediocrity) I had before whilst
doing the former type. Further, I hastily figured all my worries were quite
unfounded for this new modus operandi quickly and largely grew on
nearly each and every one of us as the workshop steadily progressed.
 10  Researcher Narratives 
   251

It was a true eye-opener for me as a teacher, researcher, and trainer

in training to have felt as though I had finally found that missing link
between practice and theory and justification of theory through practice
and not vice versa. Unlike the previous experience, I felt this time in the
thick of it all in class both theoretically and practically for my colleagues
as a mentor and my learners as a teacher researcher. We were thriving
together in solidarity and learning or helping one another to catch or
unearth substance. At the end of five weeks, I was simply ecstatic and
could not wait to go back to my own context/school to try it as a more
expert hunter-gatherer then.
After the summer break, we were back to school to try this promising
method of teacher research. Together with a colleague of mine, who had
also attended the same workshop, we set up a pilot study in our own
institution. A small group of teachers (ten teachers) from different back-
grounds and with different qualifications and levels of motivation and
teaching experience formed and experimentation began and continued
throughout the academic year through input sessions, cluster discussions,
and in-class practice and observations. Topics varied from factors affecting
classroom management, instructional delivery, and professionalism to at
times more philosophical but complex issues such as teacher and learner
education and development and educational or cognitive psychology.
Incidentally, I had a chance to continue and finalize my own EP study
I had begun during the workshop on learner involvement and agency in
curriculum and syllabus design and development. Through Potentially
Exploitable Pedagogic Activities (PEPAs) such as pair/group work, (com-
pare/contrast) essay writing and debates, together with my learners, we
tried to establish the underlying factors as to why there is often worry-
ingly very little learner say in design and development of educational
programmes. Some of the problems we listed are as follows: zero or little
training and awareness in the concept area; lack of self-esteem/confi-
dence; validity and reliability issues; and personal reservations regarding
determination and appointment of authority and control.
Though they were not asked to come up with any ideas, learners also
produced ideas to tackle the very issue; most of them being of highly
political nature, to the most extreme extent such as overhauling alto-
gether the school management systems and facilitating and/or increasing
252  Developing Language Teacher Autonomy through Action Research

wider and stronger student representation in school academic boards as

well as organizing awareness and skill raising courses, seminars, or work-
shops. It was really rewarding to see my learners entrenched in their own
learning this much and establishment of and finding solutions to prob-
lems that concern them more than anybody else. I guess they were sur-
prised and liked it all too because they said they felt incredibly capable of
taking ownership of their own learning, for a number of them a very first.
Freedom to choose anything as foci, the down-to-earth nature of
enquiry, non-flamboyant application, and a perfect marriage between
implementation of a small-scale research study and normal in-class peda-
gogic activities was simply uplifting for everyone and easy to manage
in terms of actual desired outcomes yet time and again it was also hard
to analyse and digest with regard to further teaching/learning implica-
tions and change of practice. For instance, it was so fulfilling to discover
together that the main reason behind absenteeism of some of our learners
that we thought first was because of lack of motivation was rather because
of pressure of learning they felt imposed by their instructors.
The group’s efforts and goodwill culminated in presentations at an
international English Language Teaching (ELT) conference we held in
partnership with a renowned British language and teacher-training acad-
emy. Some of these instructors are now in the writing-up process of the
conference proceedings to be published next year and others have chosen
to try their luck elsewhere at other prestigious ELT conferences and/or
publications. We all agree that we are definitely doing it again next year.

 nglish Language Teachers’ Research Reticence:

E
A Multi-Case Study

Rukiye Eryılmaz

Denise: Teacher Research Is Like a Cactus

I am an experienced teacher, and I have taught English for 14 years in

various institutions. As a CELTA-certified teacher and native speaker of
English, I have taught general English, business English, exam prepara-
 10  Researcher Narratives 
   253

tion, and English for Specific Purposes (ESP)/EAP classes to pre-school

children, teenagers, and adults across all competency levels. I have 12
teaching certificates obtained from several courses. I am also responsible
for checking the validity of the exams that are in use at the preparatory
school where I am currently working.
I strongly believe that teacher research is for those who have an inter-
est in an academic career and additionally a suitable schedule to carry out
these academic studies. In addition, I disagree with the assumption that
TR contributes to teaching practices because I believe that I have enough
skills already to be a good teacher. Though, I admit that attending in-house
trainings, seminars, and workshops are important factors that contribute to
teaching practices, I do not believe in the necessity of conducting research
since I find it very academic and I draw a line between being a teacher and
a researcher. In relation to this, that there should be some motives that
trigger carrying out action research such as doing an MA or PhD or some
career development because that is why most people do it.
Furthermore, the quality of research carried out in our school is not
good owing to the fact that teachers do not have enough time to con-
duct research. This is also the reason why I do not want to attend the
conferences in which these studies are presented or the research publica-
tions read. Although I carried out two teacher research studies—both
collaboratively—I do not want to be engaged in any research projects
anymore. I only carried out these studies because they were mandatory at
the institution where I worked. As a strong believer of making one’s own
decisions, I am annoyed by this compulsivity. I don’t think somebody has
the right to tell you that you must go away and do research on a topic.
If you want to actually specifically go ahead and write something, that is
OK. I think that depends on the person.
Another disengaging factor for me was the lack of guidance and the
chaos that emerged from this. My partner and I were not guided clearly
and we chose the topic, which ended up as an issue, because somebody
else had chosen the same topic. This incident indeed frustrated me and
became a critical turning point in my TR engagement because after
that I took a long break from carrying out research. The problem we
went through was not resolved and resulted in my disengagement from
research.
254  Developing Language Teacher Autonomy through Action Research

I was also resentful when asked to write my research to be published in

a book because I had not carried out my research to be published. I did
it with the purpose of benefiting from it in my classroom practices and
this was why I chose a certain topic that I wanted to learn more about.
As mentioned above, not having enough time to do research is another
deterrent factor. In order to do research one has to find time and usually
this means working at home as well. I am strictly against allocating time
for teacher research outside school because we need personal time outside
work.
I have found  action research to be like a cactus: if you touch it, it
hurts you. You try to achieve something but each time you try it gives
you more pain. I find it painful and do not hold any other good feelings
towards action research.

Leyla: Black Clothes on a Sunny Day

Holding a BA degree in ELT, I have been working as an English teacher

for seven years. I first engaged in research in 2010 in collaboration with
one of my colleagues. In 2013, I conducted a research project on my
own. After that time, I lost interest in research due to several reasons.
I view action research as a mandatory activity which is useless and non-­
practical. Since it was mandatory, I found myself engaged in it. However,
I never really felt interested in it. This was firstly because AR was an
obligation imposed by the administration and teacher trainers. I think
the administration saw AR as a tool to be professional; however, I do not
think that AR can provide this. Overall, I do not consider it beneficial,
and as a result I do not have the urge to do research. The fact that the
research should be made public also annoyed me in the sense that I did
not want to present my research to others. In fact, this is the main reason
for my reticence towards research, as I perceive it to be personal, not to
be publicly shared.
Another factor that caused my disinterest in research was lack of knowl-
edge on how to do it. They just told us, “You have to do it, you have to
do it”, but they did not tell us how to do it. Due to this misguidance, my
partner and I prepared our first research in only a week, which resulted
 10  Researcher Narratives 
   255

in a crisis between us and the research mentor. We were accused of being

inattentive and lazy! Following this disappointment, I did not engage in
research for some time until I embarked on another research project in
2013. This time, I took all the responsibility of my research myself and
spent more time on it because I did not have a partner.
I re-engaged in research in 2015 as my institution officially declared
the necessity of taking part in research. We were given a choice of action
research, EP, and lesson study, and I preferred to join the EP group. This
time I felt relatively more engaged since I found EP more practical com-
pared to other types of research and I was pleased to carry out a study in
a classroom environment. Besides, I am currently happier about the fact
that I have more options to choose from.
The students always force us to learn something new, and the main fac-
tor that improves you as a teacher is not research but students. I criticize
myself through the eyes of my students. In addition to practical teaching
in the classroom, I believe in the power of searching on the Internet.
When I think I lack certain information related to teaching, I simply look
for some practical answers online. Indeed, I do not find myself experi-
enced enough, but more interaction with my students in class will help
me to be more skilful in teaching. I will not achieve this via research
because it is also boring and useless. It is like wearing black clothes in a
sunny day: it is gloomy and uncomfortable and makes me feel depressed.

Martin: Tasteless, but You Have to Eat

I have worked as an English teacher since 2001, and hold a degree in fine
arts. I run my own graphics/design studio, so have blended my artistic
side with my teaching practices, as well as in my research engagement.
For instance, I preferred to do research on similar issues such as the use of
a smart-board or the efficiency of the material design.
I see action research as a mandatory activity imposed by the admin-
istration just to make sure that the teachers are working properly. In
fact, I connect AR to the distrust between the management and the
teachers. The administrators invented the necessity of carrying out
research assuming that teachers need more development and do not
256  Developing Language Teacher Autonomy through Action Research

work enough. Moreover, it is a tool by which the institution promotes

itself and gains reputation. Research is too academic in a scientific way,
and teacher researchers do it for self-promotion—to make people hear
their voices.
AR is great for those who want to, but I am not one of them. I do not
want to do research mainly because of the fact that it does not contribute
to classroom practices. Research is what people in the sciences do. It is
different than teaching. We are losing our focus. Our purpose is teaching
English. AR is not our priority.
The incident that triggered my research reticence happened when my
research mentor disapproved of the research topic I wished to work on
because someone else had chosen the same topic. Furthermore, the grad-
ual institutionalization of AR as a complementary part of my teaching
duty increased his opposition. When I first engaged in research in 2010,
it was optional. However, in time it became mandatory and demanding.
The teachers were not only required to do research but they also had to
present it every year. Also, I am not fond of engaging with other teachers’
research.
AR is like the food at the school cafeteria: it is tasteless, but you have
to eat.

Meta-Narratives
 eflection Through Written Stories of Experience
R
in a Teacher Education Course

Gary Barkhuizen
University of Auckland
I am a firm believer in reflective practice, in other words, thinking
deeply about what one does in the classroom in order to understand prac-
tice and then change it for the better and change oneself in the longer
term. I am particularly excited when that reflection takes the form of
storytelling. This is because I believe that narrative is one of the ways
in which we make meaning of our lives. In constructing stories, either
spoken or in writing or in multimodal forms, we share interpretations of
 10  Researcher Narratives 
   257

our experiences with others and in the process co-construct new stories,
leading to new interpretations. It is easy to see how this storytelling pro-
cess would be beneficial for teachers. In my language teacher education
classes at university I regularly ask my students (both pre-service and in-­
service teachers) to reflect on the content of what we doing in the class in
relation to their own experiences as language teachers and also language
learners.
In one particular class, Sociolinguistics and Language Teaching, the
teachers each produce a series of four written personal narratives over the
course of the semester to relate what we do in class to their personal his-
tories and imagined futures. For one of the course assignments the teach-
ers analyse their narratives in order to explore socially situated thematic
threads in the content of the narratives and to further make meaning of
their personal teaching experiences, both past and future. Their analy-
sis is integrated with the appropriate theoretical literature and concepts
encountered in the course.
I had been teaching the course for a few years and on each occasion
the teachers were required to reflect narratively in this way. After a while I
wondered if this mode of reflective practice really was useful for the teach-
ers, and I had a number of other questions: Were they coping adequately
with reflective writing? Were they able to “story” their experiences—write
about their experiences in story form? Did they have an understanding of
narrative and the purpose of narrative writing? What were their emotional
responses to engaging with this narrative process? Were the teachers actu-
ally learning something through reflecting narratively: about themselves as
teachers, about the contexts in which they worked, and about their prac-
tice? Over the years teachers on the course had often remarked that they
enjoyed writing the narratives and that it was a meaningful part of the
course for them. Some indicated that they really did learn something new
about teaching and about themselves as teachers. I was relieved to hear such
comments, but felt I needed a more systematic measure of the success of
narrative reflective writing in his course, especially since I was also aware
that some teachers did not perceive the writing experience so favourably. I
wondered if it were possible to do so quantitatively, and so decided the next
time I taught the course to “experiment” using a questionnaire approach, a
258  Developing Language Teacher Autonomy through Action Research

pre- and a post-questionnaire, distributed at the start and end of the course,
respectively.
There were 21 teachers in the class and all submitted both versions of
the questionnaire. Ten questions were included in both questionnaires,
each set a reflection of the other. The pre-questionnaire referred to the
beginning of the narrative writing process, eliciting expectations of the
experience, and the post-questionnaire asked for teachers’ responses after
having completed writing the four narratives. The questions covered
three broad topics: (a) the affective dimension of narrative reflective writ-
ing, particularly teachers’ levels of enjoyment, anxiety, and confidence;
(b) questions which focused more directly on narrative, particularly the
goals of narrative reflective writing and the concept of narrative itself;
and (c) topics that relate to teachers’ knowledge, the quality and effect
of reflection, change as a result of reflection, learning about teaching,
and the benefits of narrative reflective writing. Each question required
a quantitative response on a five-point Likert-type scale, with the more
positive responses on the point 5 end of the scale. The questionnaire also
allowed an open-ended comment indicating a reason for the numerical
choice. Teachers were strongly encouraged to write a comment for each
question, and the response to this request was excellent. In short, the
aim was to investigate any differences between the teachers’ knowledge
and experience of narrative reflective writing at the start of the 14-week
course and at the end.
Although the difference between the responses to the pre- and post-­
questionnaire was statistically significant (an inferential statistical analysis
was carried out on the questionnaire data), and the comments categori-
cally reflect this difference, the teachers’ quantitative choices were evenly
split among options on the positive response end of the Likert scale—
points 3, 4, and 5 (i.e. they did not all choose option 5). This perhaps sig-
nals some sort of ambivalence on the part of the teachers, even at the end
of the course. One teacher pointed out that “I understand better than
before”, suggesting perhaps that she still has a lot to learn, and another
teacher said that she “still needs more practice at analysis”, that is, inter-
preting and making meaning from her writing.
Moving to the main question which my action research aimed to
answer, which is concerned with the effects of actually writing narra-
 10  Researcher Narratives 
   259

tives on teachers’ dispositions and skills, the data seem to suggest that the
effects were positive. In general, the teachers reported that they enjoyed
writing narratives more than they had expected, that they were less anx-
ious and more confident when they wrote the last narrative than when
they wrote the first, and that they knew more about narratives and under-
stood their purpose in the course better. In a sense, this is not surprising,
because we would expect more positive results on all of these items after
the teachers had experienced narrative writing in the course. In addition,
the move towards positive trends overall should also be considered in
light of the support that the teachers received during the course, a lecture
and readings on narrative, as well as the guiding, non-evaluative com-
ments I provided each time they wrote their narratives (both to the whole
class and individually in writing on their narratives). In comparison with
the post-questionnaire results, the more negative pre-questionnaire could
also be interpreted in terms of fears of the unknown, which for most of
the teachers dissipated once the task was completed. The fact remains,
however, that the teachers’ attitudes and their perceptions of their ability
in and knowledge of narrative changed during the course (apart from
the one or two “outliers” who still responded negatively to some of the
questions at the end of the course). In other words, if narrative does
come naturally to teachers, narrative writing and the use of narrative to
­understand experiences of teaching is a craft that can be worked upon
and improved. I will certainly continue to ask my student teachers to
share with me and their classmates their stories of life and teaching expe-
riences in the classes I teach.

 ction Research, Language Teacher Education,

A
and the Construct of Investment

Bonny Norton
University of British Columbia, Canada
It was over two decades ago that I developed the construct of “invest-
ment” as a complement to the construct of motivation (Norton Peirce
1995; Norton 2013). What I had observed in my early research is that a
language learner can be highly motivated in one classroom, but passive
260  Developing Language Teacher Autonomy through Action Research

and sometimes resistant in another classroom. On closer analysis, I learnt

that if a language learner experiences marginalizing practices such as rac-
ism, sexism, or homophobia in a given classroom, she or he is likely to
withdraw from productive classroom activities, and appear a “poor” and
unmotivated language learner.
It was for these reasons that I developed the construct of “investment”,
arguing that a student can be a highly motivated language learner, but
not invested in the language practices of a given classroom. Inspired by
the work of Pierre Bourdieu, whose analysis of language and power is
compelling (Bourdieu 1991), I developed the construct of investment to
signal the relationship of a language learner to a given target language,
in the context of prevailing social, economic, and historical conditions
in both classrooms and communities. In addition to asking, “is a learner
motivated to learn a language”, I have argued that it is equally produc-
tive for a teacher or researcher to ask, “is this learner invested in the
language practices of a given classroom or community?” Most recently, I
have worked with Ron Darvin to provide a more comprehensive model
of investment, which locates it at the intersection of identity, capital, and
ideology (Darvin and Norton 2015).
In my language teacher education classrooms at the University of
British Columbia in Vancouver, Canada, I have discussed the construct
of investment with language teachers, many of whom wish to better
understand the language learners in their own classrooms. In 2012, there
was a particularly memorable event in one of my teacher education class-
rooms, discussed more fully in Norton (2016). The question I was grap-
pling with was as follows: If teachers are to promote learner investment
in the language practices of their classrooms, what implication does this
have for language teacher identity?
Keeley Ryan (2012), one of the students in my class, was struggling at
that time with great attrition in her adult ESL classrooms in Vancouver.
She decided therefore to conduct a small action research study with her
language learners, to investigate why there was attrition in her classes. In
a term paper (Ryan 2012), Ryan compared two adult education classes
(let’s call them Class X and Class Y), before and after she had read a con-
siderable body of research on identity, language learning, and investment.
In Ryan’s first adult education class (Class X), taught before Ryan was
 10  Researcher Narratives 
   261

familiar with theories of identity and investment, she noted that the num-
ber of students dropped from 25 to 9 over a period of three months. As
she reflects, “I asked my colleagues what they thought I should do about
the situation. I was told that this phenomenon was quite common, that
my students were just not motivated to complete the course” (2012, p. 4).
Ryan was not satisfied with this explanation, and in her second adult
education class (Class Y), after reading extensively about learner invest-
ment, she decided to adopt a very different set of practices in her teach-
ing. She began the class with a comprehensive questionnaire in which the
students provided information on their experiences of learning English,
their expectations of the class, and their hopes for the future. As Ryan
learnt more about the students, not only from the questionnaire, but also
from careful observations of classroom activities, she altered her practice
“to reflect what [she] imagined their idea of a good school would look
like” (p. 6). In contrast with Class X, she found very little attrition in
Class Y, with 25 of 29 students remaining till the end of the class. As
Ryan explains, “I tried to match the subject positions of the majority
of the class, as expressed through the questionnaire, with the language
practices of the classroom, in order to maximize the investment of the
majority of the students.” Ryan emerged from Class Y with a greater sense
of accomplishment and legitimacy as a language teacher.
Although Ryan’s action research study was small, and certainly not
statistically significant, it does suggest that the construct of investment
resonates well with language teachers. Also of interest is that Ryan’s action
research study in my teacher education class also had a positive impact on
my own identity as a language teacher educator and scholar.

References

Bourdieu, P. (1991). Language and symbolic power. Boston, MA: Harvard

University Press.
Darvin, R., & Norton, B. (2015). Identity and a model of investment in applied
linguistics. Annual Review of Applied Linguistics, 35, 35–56.
Norton, B. (2013). Identity and language learning: Extending the conversation
(2nd ed.). Bristol: Multilingual Matters.
262  Developing Language Teacher Autonomy through Action Research

Norton, B. (2016). Learner investment and language teacher identity. In

G.  Barkhuizen (Ed.), Reflections on language teacher ıdentity. New  York:
Routledge.
Norton Peirce, B. (1995). Social identity, investment, and language learning.
TESOL Quarterly, 29(1), 9–31.
Ryan, K. (2012). Identity and investment: Issues in an adult education classroom.
Unpublished paper for LLED 510, University of British Columbia, Canada.

Question to Consider
Which of these narratives do you most identify with?

Task
1. Identify the various themes in these narratives (open coding).
2. Organize the themes into groups (axial coding).
3. Select an overall theme (selective coding).
4. Write a narrative of your own.
 11
Conclusion

There are several ways of engaging in research to make meaningful and

tangible progress in teaching and understanding as a teacher. Different
methodologies can be followed to investigate practices, beliefs, and
knowledge such as qualitative and quantitative or mixed methodologies
by using a wide range of tools to generate data with students in spe-
cific contexts. Research can be initiated for what is not functioning well
enough or what is functioning well. Deeper insights into classroom prac-
tices, and own understandings through finding evidence is what helps
teachers develop professionally.
This book has defined action research (AR), and discussed different
types and stages of AR. It has suggested ways to clarify research questions
and to decide on appropriate research methods. It has provided both
theoretical and practical guidance for collecting, analysing, and discuss-
ing the data, as well as for presenting and writing up the results. It has
discussed important issues such as collaboration, research quality, fea-
sibility, context, triangulation, piloting, and ethical questions. In addi-
tion, it has introduced a wide range of opportunities for investigation and
understanding professional practices through tasks, thought-provoking
questions, and examples.

© The Author(s) 2017 263

K. Dikilitaş, C. Griffiths, Developing Language Teacher Autonomy
through Action Research, DOI 10.1007/978-3-319-50739-2_11
264  Developing Language Teacher Autonomy through Action Research

We have aimed to help teachers develop autonomy by offering a sys-

tematic engagement in research stages integrated with practical issues
of teaching. Autonomy in doing AR and in teaching are interrelated
processes in that the former could provide enlightenment and stimu-
lation to take control of the teaching and learning situation, whereas
the latter—autonomy in teaching—could provide an intellectual basis
which encourages research for professional development. There are sev-
eral studies that report the potential relationship between doing AR and
developing teacher autonomy (Lieberman 1995; Wang and Zang 2014).
Developing better understanding of classroom practices through doing
AR paves the way to increasing self-confidence and maturing compe-
tence to initiate and sustain professional development through inquiry-­
oriented procedures.

The Development of Teacher Autonomy

An ongoing theme throughout this book has been the development
of teacher autonomy. By engaging in AR, teachers might go through
changes and/or developments in their teaching practices and the ways
in which they conceive teaching and learning, which could be cognitive,
practical, or interactional. These kinds of impact areas enabled by research
engagement on professional development can empower teachers consid-
erably and contribute to a sense of confidence and autonomy in different
perspectives, including the following:

Cognitive Impact

Cognitive impact refers to mental changes and/or developments that

might happen in the mind through engagement in AR.

• Self-confidence: With the empowerment gained throughout the

research, self-confidence may develop in understanding and dealing
with classroom issues or teaching practices.
 11 Conclusion 
   265

• Motivation: Learning and understanding through researching could

also lead to more motivation to teach and learn as a teacher.
• Change or growth in beliefs: Beliefs might also change or develop as a
result of illumination gained. For example, teachers who resort to
drills in the classroom very often might realize that they help stu-
dents only to a limited extent and other ways should be found of
practising language skills. Seeing that drills are of little use, beliefs
about their use tend to change. Beliefs and practices are two concepts
that can complement each other. They develop or change by mutual
impact.
• Knowledge increase: During AR, researchers also read papers or books
about the issue under investigation. In the process, the terminology or
jargon of applied linguistics and theoretical knowledge are also learnt.
For example, an action researcher who investigates corrective feedback
will definitely learn the types of error and error correction and the
associated terms. Such research reading engagement will expand con-
tent knowledge to varying degrees.
• Attitudinal change: Attitudes can be negative, neutral, or positive. As a
result of research engagement, attitudes towards particular issues may
change. For example, while negative attitudes may be held towards the
use of some teaching techniques, attitudes might change from negative
to positive as a result of knowledge and practice.
• Awareness raising: Awareness is the realization of a different dimension
in what is being learnt. Awareness is one of the facilitative factors for
learning to take place. Without becoming aware of particular things,
learning may not commence. Investigating might lead to critical real-
ization about several different aspects in the classroom, such as aware-
ness of students’ preferences, styles, and strategies, the effectiveness of
new instructional practices, and unexplored factors that hinder or
facilitate learning.
• Widening perspectives: Teachers might also widen their horizons when
they engage with different facets of knowledge and understandings
through reading research and doing research. They might start to look
at pedagogical issues from different perspectives, which might influ-
ence how they teach.
266  Developing Language Teacher Autonomy through Action Research

Practical Impact

Practical impact refers to the changes and/or developments in what teach-

ers do in the classroom and how classroom issues are dealt with once AR
has commenced.

• Using new teaching practices: One of the impact areas of engagement in

AR on teachers is changes or modifications in teaching and instruc-
tional practices. Having tried out new practices and reported positive
influence of learning and learners, they might be practised in the future.
• Using existing materials differently: Research undertaken might give dif-
ferent ideas about how else the same materials might be used. Rather
than change materials (also an option), more effective ways of using
them for different purposes might be found.
• Evaluating materials with a different approach: Growing knowledge and
experience in the research process might enable evaluation of the mate-
rials available with a different eye with regard to their effectiveness,
clarity, appropriateness, and so on.
• Generating different materials: Informed by the AR, teachers could
think about what kind of materials can help achieve goals and improve
teaching plans. They can start to generate their own materials that
would address their own needs.
• Modifying instructional decisions: Though not easily observed, every act
of teaching is preceded by a set of decisions taken by the teacher. With
the personal and professional growth and development brought about
by research involvement, action researchers might reconsider instruc-
tional decisions.

Interactional Impact

Research is a social activity that brings together researchers, teachers, and

learners or others. Therefore, it is not surprising that the research process
influences relationships. Interactional impact refers to how engagement
in AR helps develop relationships in the immediate context and in the
wider community.
 11 Conclusion 
   267

• Teacher–learner relationship: Research in the classroom requires inter-

action especially when qualitative research is conducted. Teachers, as
researchers, initiate dialogues with learners to elicit their views, com-
ments, and opinions on a particular issue under investigation. This
leads to better mutual understanding which might help develop learn-
ing and teaching.
• Teacher–teacher collegiality: AR is conducted in a context consisting of
different people working together. Any research done in a context
could provide implications for others too. In this sense, sharing and
discussion of research with other teachers could encourage collegiality
with others.
• Teacher–administration bonds: AR could also bring together the admin-
istrative staff and action researchers. Research carried out in a context
is important to school development in that it might specially influence
curriculum and top-down decisions on pedagogical practices through
the local knowledge generated by research. In this sense, action
researchers could function as the internal programme evaluators who
can provide insider perspectives for improving the effectiveness in the
implementation of the curriculum of the schools.
• Teacher–colleague and researcher networking: Doing AR might stimulate
the presentation and writing of research to disseminate it with a wider
community. Sharing research with other people could initiate intellec-
tual discussions and ongoing relationships. By learning with and from
others in social settings, teachers develop interactional abilities, which
is another way of sustaining positive effects of AR in the profession.

Guidelines for Autonomy

There are several ways of developing autonomy in professional life, includ-

ing addressing some of the types of validities suggested by Hendricks
(2009, pp. 113–115).
Outcome validity: how will I use the results?
Outcome validity refers to how much the research is able to increase
understanding. This type of validity can be achieved by sharing the results
268  Developing Language Teacher Autonomy through Action Research

through presentation or publication, which might provide ongoing

thinking and planning.
Catalytic validity: are my views or practices changed?
Catalytic validity refers to resolving problems, and transforming prac-
tices. More specifically, it means how much AR functioned as a catalyst
that influenced teaching practices. This type of validity can be achieved by
critical and proactive reflection upon research engagement and meaning-­
making processes.

Sustainability
In addition to the concept of teacher autonomy, we have also high-
lighted the role of sustainability in teacher professional development.
Sustainability is what happens to the action researcher after the initiation
of research. We mean how action researchers continue to benefit from
the positive and meaningful impact of AR in their teaching careers. As
AR is a professional development tool that is empowering (Borg 2015),
the during—and after—effects could lead to thinking more deeply about
practices in the classroom. When compared to short-term professional
development activities, AR requires longer engagement as well as inten-
sified focus on practical issues though developing an investigative lens.
In this sense, it leads to becoming continuously engaged in using and
developing insights into teaching. The degree of sustainability is then
an indicator of the effectiveness and usefulness of AR as a professional
development activity. In this book, by explicitly providing ways of how to
conduct AR, we offer ways of extending the effects. The following ques-
tions might be asked:
What action have I taken?
Teachers need to think about the actions already taken or which might
be taken in the future. This creates a powerful bridge between research
and practices. Asking this question explicitly can raise awareness into
practical outcomes of the research.
 11 Conclusion 
   269

How has AR addressed my puzzles, questions, or issues?

This question might help develop a direct link between what prompted
this AR and how the research developed understanding of these issues.
Trying to understand the links between theoretical and practical aspects
contributes to seeking new meanings and actions.
How will I monitor the effects of my practice?
Monitoring the effects of practice is part of evaluating the sustaining
impact of research on practice. This question might lead to taking further
advantages of the research at a practical level.
What links are there between the issues problematized in the research and
the ways I can relate them to my classroom implementations?
AR is seen as one of the effective ways of developing teaching practices
and instruction as well as awareness towards, beliefs about, and knowl-
edge of pedagogy. Doing AR could also activate several unobservable and
unpredictable influences on teachers, which could lead to growth in self-­
efficacy, and gradual development in cognition. Teachers who become
the generators of their own pedagogical knowledge acquire ability to
question their practices, and to understand why their teaching practices
work or don’t work. Understanding why things work or don’t work will
help them make informed decisions and develop control or agency over
what and how they teach. By acquiring critical knowledge about the
applicability of their own practices through researching, teachers develop
a sense of intuition or judgement about how and why to do things in a
particular way.

Benefits of AR
AR helps to empower teachers and helps them to develop agency for
gaining pedagogical insights into potential practical teaching challenges.
In the literature the benefits of AR are closely related to teachers tak-
ing control of their own teaching practices. Engagement in research and
development of autonomy support each other in that the former serves
to develop the latter.
270  Developing Language Teacher Autonomy through Action Research

AR can be liberating in that it can lead to emancipation from the

dependence on expert knowledge and develop the ability to question our
existing ideas and practices as well as creating our own way of doing
things in the classroom. Liberated teachers are those who hold the con-
trol of their knowledge and understanding of their own practices and
classroom situations. Positive benefits gained by AR engagement are criti-
cally sustained and free from superficial examination of issues and ques-
tions in favour of deeper research-based elaboration. Liberation brings
teachers more confidence to deal with pedagogical issues.
The authors of this book hope that it may empower teachers to initiate
and sustain their research experience and benefit from research as a liber-
ating professional development activity which leads to teacher autonomy.

Questions to Consider
1. How did you start doing AR and why?
2. What challenges did you experience at the initial stages, and how did
you overcome these or did you ever?
3. What were the facilitative factors that helped during your research
experience?
4. What strategies did you follow?
5. Can you write some anecdotal description of interesting and critical
incidents that occurred?
6. Did you present your research somewhere? If so where and how did it
go? Did you receive any feedback that provided you different insights?
7. How long did it take you to complete your research? Was it completed
in the planned period?
8. Did you publish your research? If yes, where?
9. What were your writing experiences? How did the writing experience
influence your understanding of issues raised in the research?
10. What would you like to reflect upon as benefits you gained from the
research experience?

Task
By filling out the following checklist, you can evaluate yourself as a
researcher and understand to what extent you have developed knowledge
about AR principles.
Checklist for Teacher Perceptions of Action Research
 11 Conclusion 
   271

Please rate each statement according to

5=strongly agree 4=agree 3=neutral 2=disagree 1=strongly disagree

Statements Rating
1. I expect to benefit professionally from my research
experience
2. I expect to benefit academically from my research
experience
3. I expect to benefit personally from my research
experience
4. I expect my students to benefit from participating in
research
5. The research experience has improved my relationship
with my students
6. I intend to do more action research
7. My institution was supportive of my research
8. I feel my autonomy as a teacher has increased as a result
of my research experience

Final Task
1. Write up a complete version of the study you have been working on
throughout this book.
2. Find ways of disseminating your research (e. g. at a staff meeting or
conference, online, in a newsletter or journal, or as a poster in an
event).
 Appendix: A Mixed Methods Study

Classroom Culture
Carol Griffiths

The small-scale action research study on classroom culture which has been
progressively used to illustrate the various steps throughout this book is
an example of a mixed methods study in that it uses a questionnaire to
gather quantitative data about student perceptions of impoliteness in a
university classroom and also a grounded approach to qualitatively anal-
yse the comments. Since the piecemeal presentation to which it has been
subjected chapter by chapter may well have destroyed any sense of how
it “hangs together”, the full version is presented here for those who may
be interested. Note that, since part of its intended purpose was to act as
a model for how the students might go about their own research, there
is no abstract, keywords, or introduction, and only a very brief litera-
ture review, all of which would need to be added or extended if it were
intended for submission to a journal (which was never the case with this
study).

© The Author(s) 2017 273

K. Dikilitaş, C. Griffiths, Developing Language Teacher Autonomy
through Action Research, DOI 10.1007/978-3-319-50739-2
274  Appendix: A Mixed Methods Study

Literature Review

Culture is always a difficult concept to define and is particularly hard

to separate from concepts such as ethnicity and nationality (Zhou and
Griffiths 2011). In its anthropological sense, culture is regarded as any of
the customs, worldview, language, kinship systems, social organization,
and other taken-for-granted day-to-day practices of a people which set
them apart as a distinctive group (Scollon and Scollon 2000), whereas
ethnicity refers to race (e.g. Maori) and nationality is essentially a politi-
cal concept (what goes on one’s passport).
A classroom is a specific sociocultural environment which has its own
set of rules, language, expected behaviour, and standards of politeness
(e.g. Griffiths et al. 2011). If these standards are not met, it may stop us
from working well with others whose norms regarding what is polite or
appropriate are different from our own (e.g. Neuliep 2003). It is therefore
important to be clear about the behaviour expected, especially since these
behaviours might not be identical across all contexts.

Research Question

This study asked the following questions:

What are the expected standards of behaviour in a Turkish university
classroom?
Do these standards vary according to gender or nationality?

Research Context

Participants

There were 35 students who participated in this study. They were in the
fourth year of a four-year degree in English Language Teaching (ELT),
so they had already been at the university for some time, and might,
therefore, have been expected to have clear ideas about the standard of
behaviour expected. There were 13 male students and 22 female. Turkish
students were a majority (N = 24), and there were 11 international stu-
   Appendix: A Mixed Methods Study 
   275

dents (from Nigeria, Turkmenistan, Tajikistan, Iraq, Kirgizstan, Greece,

and Kazakhstan).

Setting

The students were attending a Sociolinguistics course in the ELT

Department in the Education Faculty at a private university in Istanbul,
Turkey. Since the study was carried out in an actual classroom, where stu-
dents were mainly concerned with passing their course (especially since
many hoped this would be their last semester before graduation), the
study had a multiple purpose: in addition to the research purpose, the
topic was used to stimulate discussion, and also to act as a model for
research projects in which students were engaged as part of their course.

Data Collection

As a first step, the issue of politeness was discussed in class, and students
were asked to provide ideas about their perceptions of polite and impolite
behaviour. They wrote these ideas on pieces of paper which were handed
in anonymously. Based on this input, a questionnaire was then constructed
(see Appendix for the questionnaire). A five-point Likert scale format was
used, ranging from 5 = very impolite to 1 = not at all impolite. In addi-
tion, in order to obtain qualitative data, a column was provided where stu-
dents were asked to comment and explain their opinions. In the interests
of preserving privacy if students chose to remain anonymous, providing
their names was optional, and in order to ensure consensual participation,
students were asked to sign an agreement to using the data for research or
publication purposes at the bottom of the questionnaire form.

Data Analysis

The questionnaire ratings were entered into the Statistical Package for
Social Sciences (SPSS) and, since Likert-scale data are nonparametric,
the data were analysed for median ratings. In addition, Mann–Whitney
276  Appendix: A Mixed Methods Study

U tests of difference for nonparametric data were used to investigate

any differences in the ratings according to gender or nationality. The
comments were also examined for ınsights which helped to explain the
ratings.

Results

The median ratings results are set out in Table 1. According to these
results, Items 7 and 8 are rated 5 (“very impolite”), whereas none of the
items are rated 2 or 1 (in the “not impolite” range)
According to a Mann–Whitney U nonparametric test of difference,
there were no statistically significant differences according to gender.
According to nationality, only Item 1 showed a significant difference
(p = .029), with international students considering coming late to class
significantly more impolite than Turkish students.
A selection of the qualitative results from the comments are as
follows:

1. Coming late

• Being on time is your responsibility.

• Causes lack of concentration.
• If it is temporary OK, but if they keep coming late it is rude.

Table 1  Median ratings of questionnaire items (the higher the rating, the more
impolite the behaviour is considered to be)
Item Behaviour Rating
1 Coming late 3
2 Using a phone 4
3 Talking while someone else is talking 4
4 Not paying attention 3
5 Sleeping 3
6 Eating or drinking 3
7 Using inappropriate terms of address 5
8 Using bad language 5
9 Wearing inappropriate clothing 3
   Appendix: A Mixed Methods Study 
   277

2. Using a phone

• It is disrespectful.
• It disturbs others.
• OK for emergency, but it causes students to lose their
concentration

3. Talking while someone else is talking

• We should behave as we want to be treated.

• Plain rude.
• Everybody needs to respect one another.

4. Not paying attention

• This is disrespectful.
• It distracts others
• Don’t attend the class if you won’t pay attention anyway.

5. Sleeping

• There is always coffee.

• It is disrespectful.
• Bedrooms are for sleeping, not classes.

6. Eating or drinking

• Sometimes we don’t get time for lunch or breakfast.

• OK unless loud or smells bad.
• Maybe he/she has an illness.

7. Using inappropriate terms of address

• This is very rude.

• It may cause unfriendly situation or arguments.
• It can be very embarrassing, but it depends on the culture.
278  Appendix: A Mixed Methods Study

8. Using bad language

• This is impolite not just in class but everywhere.

• Can cause anger, losing concentration, and so on.
• Unrespect for the teacher.

9. Wearing inappropriate clothing

• It may be distracting. This is a school.

• It is a personal choice.
• It is important to be comfortable.

10. Any other behaviour you consider impolite

• Not bringing the right material.

• Not sitting upright.
• Chewing gum.
• Doing other work/reading in class.
• Being noisy.
• Leaving the class without permission.
• Attending to personal grooming (e.g. applying make-up, combing
hair)

Discussion

It is interesting that all of the items in the scale were rated on the medium
to very impolite side of the scale, with using inappropriate terms of address
and using bad language being considered most impolite (median = 5).
Using a phone and talking while someone else is talking are also consid-
ered impolite (median = 4). Some participants felt that, given Istanbul’s
notorious traffic, coming late was sometimes unavoidable, while eating
or drinking in class might be all right under some circumstances as long
as it did not inconvenience others, since there are days when students
start early and finish late with no breaks between classes. A reasonable
degree of tolerance seemed to be extended to standards of dress, as long
   Appendix: A Mixed Methods Study 
   279

as it was not “embarrassing” or “distracting”, and the question was also

raised by some students that it was not easy to achieve a definition of
what was “appropriate”, a point that was also made regarding terms of
address (these points were discussed in class as part of a follow-up to
the study). Several other impolite behaviours were suggested, including
failure to bring the correct material, chewing gum, and attending to per-
sonal grooming.
Another interesting feature of the results is the unanimity across
learner variables. We might have expected males and females to have
different opinions about some of these behaviours, but this did not
prove to be the case. Furthermore, although there was one statistically
significant difference according to nationality, suggesting that Turkish
students are less concerned about punctuality than their international
classmates, the remainder of the differences according to nationality
were not significant.

Implications

Once the results had been found and the study written up, it was
distributed to the class for discussion. There was some amusement
with some of the findings (e.g. about eating, drinking, or sleeping in
class), and some embarrassment from some who recognized impolite
behaviours in themselves (e.g. consistently coming late and entering
in such a way as to distract classmates, using phones during class).
In the weeks following the study, there was some evidence that stu-
dents made some effort to control their own negative behaviour (e.g.
by being more punctual, turning off phones when they arrived in
the room) and some also seemed to be more inclined to attempt to
control classmates for impolite behaviour (e.g. if they were talking
while someone else—either the teacher or one of their own classmates
during a presentation). In other words, this study seemed to have
some effect with raising students’ own awareness of the effect of their
behaviour on others, and to result in more willingness to engage in
polite behaviour, which, in turn, seemed to have a positive effect on
classroom dynamics.
280  Appendix: A Mixed Methods Study

Suggestions for Further Research

There are a number of ways in which this study might be extended:

• An interesting direction would be to give the survey to teachers and

compare their perceptions with those of students.
• Also, interviews could be conducted for a more in-depth qualitative
perspective.
• Some of the other behaviours the students considered impolite might
be added to the questionnaire.

Conclusion

Overall, there is agreement that four out of nine (44.4%) of the behav-
iours listed in the questionnaire are impolite or very impolite. Students
did not express strong opinions about the remaining five items (55.6%),
but none of the items was rated polite. Perhaps if students are made aware
of these findings it might help to promote a more harmonious classroom
atmosphere which is likely to promote more effective learning.

References

Griffiths, C., Raman, T., & Mannes, M. (2011). Have I offended you?
Appropriate nonverbal communication in the international classroom.
Journal of International Education and Business, 3, 7–29.
Neuliep, J. (2003). Intercultural communication: A contextual approach.
New York: Houghton Mifflin.
Scollon, R., & Scollon, S. (2000). Intercultural communication: A dis-
course approach. Beijing: Foreign Language Teaching and Research Press.
Zhou, C., & Griffiths, C. (2011). Intercultural communicative com-
petence. English Language and Literature Studies, 1/2, 113–122
 Appendix: Classroom Culture
Questionnaire

Gender: M F Nationality:
Please mark the following behaviour according to how impolite you
think it is. Please also add a comment explaining your opinion.
5 = very impolite 4 = impolite 3 = no strong opinion 2 = not impolite
1 = not at all impolite
Comment: please
Item Behaviour Rating explain your rating
1 Coming late
2 Using a phone
3 Talking while someone else is talking
4 Not paying attention
5 Sleeping
6 Eating or drinking
7 Using inappropriate terms of address
8 Using bad language
9 Wearing inappropriate clothing
Any other behaviour you consider impolite

I agree to the use of this data for research or publication purposes:

© The Author(s) 2017 281

K. Dikilitaş, C. Griffiths, Developing Language Teacher Autonomy
through Action Research, DOI 10.1007/978-3-319-50739-2
 Glossary

Abstract  A short summary of the main ideas and findings placed at the begin-
ning of an article.
Action research  First used by Lewin (1946), the term action research involves an
active investigation of a problem, puzzle, question, or issue by a researcher or
researchers (often the teacher/s) who is/are a participant or participants in the
given context (Burns, 2010; Lewin, 1946).
ANOVA (analysis of variance)  A statistical procedure for testing whether the dif-
ference between parametric (q.v.) variables is significant (q.v.)
APA  This refers to a style established by the American Psychological Association,
commonly used in social science publications.
Case study  This kind of research takes an in-depth look at a specific definable
unit, for example, an individual, a particular class, or a specific school.
Categorical data  Another term for nominal data (q.v.) where the data are divided
into categories.
Collaborative action research Teachers working collaboratively to do action
research with the purpose of investigating and reflecting critically on their
teaching practices.

© The Author(s) 2017 283

K. Dikilitaş, C. Griffiths, Developing Language Teacher Autonomy
through Action Research, DOI 10.1007/978-3-319-50739-2
284  Glossary

Confidence interval  The confidence interval indicates a certain degree of prob-

ability that we can have confidence that a given result is more than mere
chance. For instance:
p<.05 indicates that we can be 95% confident that the result is significantly
more than would be found just by chance. Even better, p<.01 indicates a
99% level of confidence that the result is significantly more than would
be found just by chance.
Context  The context refers to the environment (e.g. a nation, a district, a school,
a workplace, or a classroom) in which a given phenomenon occurs. Context
may include both the setting where the phenomenon occurs and the people
involved.
Continuous data  Operating over a range, for example, test scores from 0% to
100%.
Convenience sampling  This refers to a group of participants chosen for a study
based on their convenient availability. For instance, if I want to investigate
the effect of strategy training, I might decide to study my own class on the
basis that they are easily available, whereas other groups might not be so
convenient.
Correlation  This refers to a statistical procedure to measure the strength of the
relationship between variables. Correlations can vary between 1 (meaning
there is a perfect correlation, typically found only between items that actu-
ally are the same), 0 (which means there is no correlation), and −1 (meaning
that the correlation is negative, so that as one variable increases the other
decreases). Two well-known tests of correlation are:
• Pearson product–moment correlation or Pearson’s r for numerical (q.v.),
normally distributed (q.v.) data such as test scores etc.
• Spearman’s rank–order correlation or Spearman’s rho (rs) for data
which is not normally distributed (q.v.) or ordinal (q.v.) data such as
obtained from Likert-type questionnaires (q.v).
Criticality 
The practice of ‘socially situated reflection and evaluation’ (Banegas,
2016, p. 1) which involves creating multiplicity in perspectives while elabo-
rating on issues, including self-criticism.
Cross-sectional  These kinds of studies look at the relationships between/among
various factors at a particular point in time (cf. longitudinal).
Deductive analysis  Pre-setting the potential themes that are to be deduced from
the data.
  Glossary 
   285

Descriptive statistics This term is used to include procedures such as mean,

median, mode, frequencies, and percentages which are used to describe the
results of a particular study.
Differences  Whereas tests of correlation test how variables relate to each other,
tests of difference examine whether given variables (e.g. reported strategy use)
vary according to different groups (e.g. male/female). Common tests of dif-
ference include:
• Student’s t-test—named after William Gosset who introduced the pro-
cedure under the pseudonym “Student”, this test is used for numerical
(q.v.), normally distributed (q.v.) data such as finding the differences in
test scores according to gender.
• Mann–Whitney U—this test is used for two independent groups (e.g. male/
female) when the dependent variable is either ordinal (q.v.), for example,
reported strategy use, or numerical (q.v.) but not normally distributed (q.v.).
• Kruskal–Wallis—this test is used for groups (e.g. several different nation-
alities) when the dependent variable is either ordinal (q.v.), for example,
reported strategy use, or numerical (q.v.) but not normally distributed
(q.v.). This test is often used for nonparametric (q.v.) data where an
ANOVA (q.v.) would be used for parametric (q.v.) data.
Documentation  There are numerous documents which might be kept to serve as
evidence for research studies, including:
• lesson plans
• teaching materials
• student work
• journals/diaries/blogs
• pictures
• tests and results
Effect size  This refers to the strength of one variable’s effect on another, for
example, what is the effect of motivation on target language proficiency
development.
Ethics  Ethical behaviour in research refers to a number of issues, including:

• Informed consent: participants have a right to be informed about the

purpose of the research and how the data they provide will be used.
• Voluntary participation: participants have a right to understand that
they do not have to participate, and that they can withdraw at any time if
they wish to do so.
286  Glossary

• Anonymity: participants do not have to identify themselves if they do not

wish to do so.
• Confidentiality: information given during the data collection process
must remain confidential.
• Storage of data: data obtained during the collection process must be
stored securely and kept for an approved period of time.
• Absence of threat: participants must be assured that the data they provide
will have no effect on their results or cause any other disadvantage.
• Professional distance: teachers should take care to maintain the teacher–
student relationship at all times. (See Chap. 5 for more details.)
Ethnography  Ethnographic studies examine a particular phenomenon which
occurs in a specific cultural context.
Experimental  A true experimental study involves using:

• random assignment
• control versus experimental groups
• application of some kind of intervention
• use of pre- and post-tests to examine the effect of the intervention.
Exploratory action research  A form of research requiring extensive initial discus-
sion of practice before attempting to make further practical and pedagogical
changes.
Exploratory practice  A form of research requiring exploring the context with co-­
researchers such as learners through normal pedagogical practices as research
tools.
Factor analysis  A factor analysis is used to determine whether particular items in
a data set “hang together”. There are two types of factor analysis:
• An exploratory factor analysis uses the factor analysis procedure to iden-
tify factor groupings, for example, in a newly constructed instrument.
• A confirmatory factor analysis is used to determine whether factors do,
in fact, hang together according to some pre-­determined grouping, for
example, an existing questionnaire.
Feasibility 
This refers to how practical a given research idea may be in a given
context.
Grounded theory  Grounded theory describes a research approach which does
not begin with hypotheses, but where the conclusions emerge “from the
ground up” as a result of the investigation. Typically, there are three stages of
coding involved with grounded theory:
  Glossary 
   287

• Open coding: At this stage, the data is examined for salient themes.
• Axial coding: At this stage, the themes identified in Stage 1 (which can be
numerous) are grouped around central axes.
• Selective coding: At this final stage, a single overarching theme is
identified.
Hawthorne effect  Also called observer paradox (Labov, 1972), the Hawthorne
effect (after the place where it was first documented, Landsberger, 1958)
explains the phenomenon that, although an observer may go into a situation
in order to observe natural behaviour, the very fact of being there is likely to
change normal patterns of behaviour.
Hypothesis/es  A hypothesis states the researcher’s prediction/s concerning the
finding/s that a given study is likely to arrive at. A null hypothesis (q.v) pro-
poses that there will be no difference between the groups being researched. If
a difference is found, the null hypothesis will be rejected and the alternative
hypothesis will be supported.
Inductive analysis  This involves inducing conclusions from the data rather than
deducing conclusions according to pre-set criteria.
Inferential statistics  This term is used to include procedures such as correlation,
differences, and effect size which are used to infer generalizations beyond the
immediate sample.
Instrumentation  Instrumentation refers to the materials or instruments (e.g. a
questionnaire, interview schedule, observation checklist) used in the course
of the study.
Inter-rater reliability  This is important for qualitative studies to help reduce the
risks of subjectivity if just one researcher’s assessments are recorded. Inter-
rater reliability is often expressed as a percentage, which represents the total
number of the items on which the different raters agree divided by the
total number of items identified. Any disagreements need to be resolved by
negotiation.
Interviews  The interview technique involves canvassing the opinions of an indi-
vidual or group of individuals. Interviews can be:
• Structured (where the questions are decided beforehand and only those
questions are used)
• Unstructured (where the interview proceeds according to issues which
arise spontaneously)
• Semi-structured (where a set of pre-planned questions forms the basis
of the interview, but there is some flexibility to discuss other issues or
insights as they arise)
288  Glossary

Interval data  Where values have regular intervals between them, for example,
children’s ages in years, worker’s income in dollars.
Introspection  Introspective methods involve asking participants to reflect on
their thinking processes and to report on these while they are performing a
task. Often also called think-aloud procedures (q.v.).
Keywords  The purpose of keywords is so that key topics will appear on a data-
base for someone looking for this subject. As such, they are an important aid
to citability.
Kolmogorov–Smirnov test  This test checks whether data are normally distrib-
uted around a mean (see also Shapiro–Wilk test). If the significance value
for an item is less than .05, distribution is not normal. This is important,
since parametric tests such as Pearson product–moment correlation, t-tests,
and ANOVAs (q.v.) assume normal distribution, and, if this is not the case,
nonparametric equivalents such as Spearman rank order correlation, Mann–
Whitney U or Kruskal–Wallis H (q.v.) should be used. The Kolmogorov–
Smirnov is considered more reliable for smaller numbers (less than about
100) than the Shapiro–Wilk.
Kruskal–Wallis H test of difference  This is a nonparametric (q.v.) test useful for
testing whether any differences among several data sets which are non-
numerical and/or not normally distributed are significant. It is often reck-
oned to be the nonparametric equivalent of an ANOVA (q.v.).
Lesson study  A form of reflection supported by peer observation to develop bet-
ter understanding of actual classroom settings at the moment of teaching as
well as student learning.
Likert-type questionnaire. This kind of instrument, invented by psychologist
Rensis Likert (1932), involves asking participants to rate questionnaire items
on a continuum, typically, for instance, from strongly agree to strongly dis-
agree, from always to never or some similar range which expresses opinion
or attitude.
Literature review  A summary of the existing published literature on a given
topic.
Longitudinal  Taking place over a period of time (cf. cross-sectional).
Mann–Whitney U  This is a nonparametric (q.v.) test useful for testing whether
there are any significant differences between two data sets which are non-
numerical and/or not normally distributed. It is often reckoned to be the
nonparametric equivalent of a t-test (q.v.).
MANOVA  (Multivariate analysis of variance). A statistical procedure for testing
whether the difference among multiple parametric variables is significant (cf.
ANOVA, Kruskal–Wallis H).
  Glossary 
   289

Mean The mean (also called “average”) is obtained by adding all numbers

together and dividing by the number of the numbers. So if a group of stu-
dents score 10, 8, 8, 7, 5, 4, 2 on a test, the mean is the total (=44) divided
by how many numbers (=7), so the mean=6.28.
Median  The median is the number at the midpoint of a set of numbers. In the
example above for the mean, the median is 7.
Member-checking  Asking for the feedback of the people who provided data for
the analysed data in order to increase the reliability and validity.
Methodology  The methodology refers to how a researcher went about conduct-
ing the study, and includes details about the participants, the context, how
the data were collected and analysed, the instruments that were used, ethical
procedures, and so on.
Mixed methods  A mixed method study is one which, rather than keeping to
just one method (e.g. a questionnaire or interviews), uses more than one
method (e.g. a questionnaire, followed up by interviews). The advantage of
such methods is that they provide triangulation (q.v.).
Mode  The mode of a set of numbers is the one which occurs most frequently. In
the case of the numbers exemplifying “mean” above, the mode is 8.
Narrative inquiry  This kind of methodology gathers data by means of asking
participants to narrate their experiences, feelings, opinions, and so on. These
data are then analysed by means of qualitative methods such as grounded
theory (q.v.).
Nominal/categorical data  This occurs where variables are given numbers for the
sake of convenience for entering them into a computer programme. For
instance, males=1, females=2. Or Chinese=1, Indians=2, Turkish=3, and so
on. These values are purely arbitrary, and have no actual numerical value. As
a result, they cannot be analysed by either parametric (e.g. Pearson’s product–
moment correlation) or nonparametric (e.g. Spearman’s rho) tests, though
they can be used as grouping variables in tests of difference.
Nonparametric  Nonparametric data are not normally distributed (q.v.) and often
non-­numerical. A common example of these types of data include Likert-
type questionnaires (q.v.), where the numbers actually represent strength of
opinions (e.g. “strongly disagree”) rather than real numbers.
Normal distribution  If numbers are normally distributed, they form a symmetri-
cal bell-­shaped curve when graphed. Normality of distribution can be tested
by means of a Kolmogorov–Smirnov or a Shapiro–Wilk (q.v.) on SPSS.
Null hypothesis (H0).  This is a statement that there is no correlation or differ-
ence between given variables. If a correlation or difference is found, the null
hypothesis is rejected, and the alternative hypothesis is accepted.
290  Glossary

Numerical data  Are real numbers (e.g. ages, incomes, test scores), as opposed to
ordinal (q.v.) or nominal (q.v.) data.
Observation  One of the ways of understanding practices as well as learning
about learners and teachers through direct access to the classroom. It provides
first-hand data to the researchers or other teachers.
Observer paradox  See Hawthorne effect.
Ordinal data  This kind of data is generated from procedures such as Likert-type
questionnaires. The numbers assigned to judgements such as “strongly dis-
agree” are clearly not actual numbers: they merely indicate strength of opin-
ion about a given question/statement. Furthermore, we cannot assume that
someone who assigns a “4” to a given item is twice as much in agreement
as someone who only gives it a “2”: this simply makes no sense. In other
words, this kind of data is non-numerical, and should be analysed using non-
parametric tests such as Spearman’s rho (for correlations, q.v.) and Mann–
Whitney U or Kruskal–Wallis (for differences, q.v.).
Parametric  This kind of data consists of real numbers, which operate within set
parameters. For instance, we can say that a child of 4 is half as old as a child of
8. Someone who earns $50,000 per year earns twice as much as someone who
earns $25,000. A student who gets 60% on a test gets 2/3 of what a student
who gets 90% earns. As long as this kind of data is normally distributed (q.v,)
it can be analysed using parametric tests such as Pearson’s product–moment
correlation, Student’s t-test of difference, or ANOVA (q.v.).
Participants  Participants are those who take part in a study, perhaps by answer-
ing questionnaires, attending interviews, and so on.
Pearson product–moment correlation  A statistical test for calculating the relation-
ship between two numerical variables (e.g. exam scores, age).
Phenomenology  This is the study of a specific phenomenon (e.g. unmotivation,
age or gender differences, autonomy, metacognition.) in a given context.
Piloting  The term used for trialing an instrument or research procedure in order
to identify any problems before using it with the main target participants.
Population  A group of people who share some particular characteristic (e.g.
university students, young readers, older language learners, female language
learners) from which a sample (q.v.) is drawn.
Potentially exploitable pedagogic activities (PEPAs)  Classroom activities that inte-
grate teaching practices and research purposes especially in collecting data. These
activities benefit students rather than use them only for eliciting research data.
Probability  In statistics, this term refers to the probability that a given effect
has occurred by more than just chance. In social sciences, 95% probability
(p<.05) is generally accepted as the lower threshold to conclude that a result
is statistically significant.
  Glossary 
   291

Problematize  Investigating an issue in the classroom which is not necessarily

deficient but may be developed or understood from different perspectives.
Purposive sampling  Purposive sampling occurs when participants are selected for
some particular reason, such as their gender, age, achievement level or nation-
ality (cf. convenience and random sampling).
Qualitative research  This refers to studies which result in data which cannot be
expressed in numerical terms, such as interviews, observations, case studies,
think-alouds, and so on.
Quantitative research  This refers to studies which produce data which can be
expressed as numbers, such as Likert-type questionnaires (q.v.).
Quasi-experimental  This refers to the kind of study which attempts to assess the
effects of a particular treatment, but where one or more of the conditions for
a true experiment (e.g. random assignment, q.v.) are not met.
Questionnaire  An instrument for collecting data, with open and/or closed items.
Questionnaires can be paper-and-pencil, or, increasingly commonly, they can
be distributed electronically.
Random sampling  This occurs where participants are selected for a study purely
by chance. (cf. convenience and purposive sampling).
Reflective practice  Reflecting upon classroom practices in order to evaluate their
efficiency in a systematic way.
Regression  A regression analysis is a statistical technique for using independent
variables to predict a dependent variable. For instance, if a student scores
30% on one assessment, and 40% on another, a regression analysis could be
used to predict the likelihood of him or her obtaining a passing grade (60%).
In an example such as the above where there is more than one independent
variable, it is called multiple regression.
Reliability  This refers to the consistency of the data. There are two main types:
internal and external (see Chap. 1: Quality of the data).
Replication  Replication occurs when a study which has been carried out in one
context is repeated in a different environment in order to see if similar results
are produced.
Reports  Reports are constructed by means of gathering and analysing available
documents, such as journals, blogs, class records, and so on.
Sample  A sample is a section of a larger population (q.v.) which is used (usually
for practical reasons, since surveying an entire population is often difficult if
not impossible) to gather information about a particular phenomenon. For
instance, if we want to find out about first language maintenance among
immigrant children, we might begin by surveying children at just one school,
or even just one class, and begin to draw conclusions from there.
292  Glossary

Setting  This term refers to the place or location where particular phenomena
occur.
Shapiro–Wilk test Like the Kolmogorov–Smirnov test (q.v.) this test checks
whether data are normally distributed around a mean. If the significance
value for an item is less than .05, distribution is not normal. This is impor-
tant, since parametric tests such as Pearson product–moment correlation,
t-tests, and ANOVAs (q.v.) assume normal distribution, and, if this is not
the case, nonparametric equivalents such as Spearman rank order correlation,
Mann–Whitney U or Kruskal–Wallis (q.v.) should be used. The Shapiro–
Wilk test is usually considered to be more reliable than the Kolmogorov–
Smirnov for larger numbers (100+).
Significance  The significance (sig.) figure tells us the probability that a particular
result is likely to be more than chance. In social sciences, the lower threshold
is generally accepted as 95% (p.<05).
Spearman rank order correlation  This is a statistical test for calculating the rela-
tionship between two nonparametric (q.v.) variables (e.g. expressions of
strength of agreement on a Likert-type questionnaire).
Standard deviation (SD)  A measure of the degree to which scores deviate from the
mean. If the SD is low, it indicates scores are close together. If it is large, it
means scores vary considerably.
Sustainability  Sustainability refers to creating potential longer-term impacts of
the benefits of action research on teachers.
Teacher autonomy  Teachers’ ability to make decisions on their own professional
development and create independent and interdependent opportunities for
themselves.
Teacher research  A form of research carried out by teachers in their own working
context in order to improve their practices and develop their understanding.
Think-aloud  The think-aloud technique is a form of introspection (q.v.) where
participants are asked to verbalize their thinking processes for later analysis.
Triangulation  Where different methods of data collection and analysis are used
to cross-check the findings.
t-test  This is a parametric (q.v.) statistical procedure for determining whether
any differences between numerical, normally distributed variables are signifi-
cant (q.v.).
Validity  Validity refers to the extent to which a procedure measures what it is
supposed to be measuring. There are several different types of validity, includ-
ing construct/concept, content, convergent, concurrent, criterion, external,
face, internal and predictive (see Chap. 1: Quality of the data).
 References

Allwright, D. (1997). Quality and sustainability in teacher-research. TESOL

Quarterly, 31(2), 368–370.
Allwright, D. (2000). Exploratory practice: An “appropriate methodology” for
language teacher development. In 8th IALS symposium for language teacher
educators, Edinburgh, Scotland.
Allwright, D. (2003). Exploratory practice: Rethinking practitioner research in
language teaching. Language Teaching Research, 7(2), 113–141.
Allwright, D. (2005). Developing principles for practitioner research: The case
of exploratory practice. The Modern Language Journal, 89(3), 353–366.
Allwright, D., & Hanks, J. (2009). The developing language learner. Hampshire:
Palgrave Macmillan UK.
Allwright, D., & Lenzuen, R. (1997). Exploratory practice: Work at the cultural
inglesa, Rio de Janeiro, Brazil. Language Teaching Research, 1(1), 73–79.
Ary, D., Jacobs, L.  C., Sorensen, C., & Walker, D. (2010). Introduction to
research in education. Wadsworth: Cengage Learning.
Bailey, K. (2001). Action research, teacher research, and classroom research in
language learning. In M. Celce-Murcia (Ed.), Teaching English as a second or
foreign language (3rd ed., pp. 489–498). Boston: Heinle & Heinle.
Bailey, K., Curtis, A., & Nunan, D. (2001). Pursuing professional development.
Boston: Heinle & Heinle.

© The Author(s) 2017 293

K. Dikilitaş, C. Griffiths, Developing Language Teacher Autonomy
through Action Research, DOI 10.1007/978-3-319-50739-2
294  References

Banegas, D. L., & de Castro, L. S. V. (2016). Criticality. ELT Journal, ccw048.
Barkhuizen, G. (2009). Topics, aims and constructs in English teacher research:
A Chinese case study. TESOL Quarterly, 43(1), 113–125.
Benson, P. (2000). Autonomy as a learners’ and teachers’ right. Learner Autonomy,
Teacher Autonomy: Future Directions, 3(2), 111–117.
Besser, S., & Chik, A. (2014). Narratives of second language identity amongst
young English learners in Hong Kong. ELT Journal, 68(3), 299–309.
Borg, S. (2010). Doing good quality research. JACET Journal, 50, 9–13.
Borg, S. (2011). Doing action research in English language teaching. A guide for
practitioners. ELT Journal, 65(4), 485–487.
Borg, S. (2013). Teacher research in language teaching: A critical analysis.
Cambridge: Cambridge University Press.
Borg, S. (Ed.). (2015). Professional development for English language teachers:
Perspectives from higher education in Turkey. Ankara: British Council.
Bright, B. (1996). Reflecting on ‘reflective practice’. Studies in the Education of
Adults, 28(2), 162–184.
Burns, A. (1999). Collaborative action research for English language teachers.
Cambridge: CUP.
Burns, A. (2009). Action research in second language teacher education. In The
Cambridge guide to second language teacher education (289–297). New York:
Cambridge University Press.
Burns, A. (2010). Doing action research in English language teaching: A guide for
practitioners. New York: Routledge.
Burns, A. (2015). Renewing classroom practices through collaborative action
research. In K. Dikilitaş, R. Smith, & W. Trotman (Eds.), Teacher-researchers
in sction (pp. 9–17). IATEFL: Faversham.
Burns, A., & Hood, S. (1997). Teachers’ voices 2: Teaching disparate learner
groups.
Burton, J., & Mickan, P. (1993). Teachers’ classroom research. In J.  Edge &
K. Richards (Eds.), Teachers develop teachers research (pp. 113–121). Oxford:
Heinemann.
Cajkler, W., Wood, P., Norton, J., & Pedder, D. (2013). Lesson study: Towards
a collaborative approach to learning in initial teacher education? Cambridge
Journal of Education, 43(4), 537–554.
Carr, W., & Kemmis, S. (2003). Becoming critical: Education knowledge and
action research. New York: Routledge.
Çelik, Ç., & Dikilitaş, K. (2015). Action research as a professional development
strategy. In S. Borg & S. Sanchez (Eds.), International perspectives on teacher
research (pp. 125–138). Palgrave: Basingstoke.
  References 
   295

Child, A. J., & Merrill, S. J. (2003). Professional mentors’ perceptions of the
contribution of school/HEI partnerships to professional development and
school improvement. Journal of In-Service Education, 29(2), 315–324.
Cochran-Smith, M., & Lytle, S. L. (1999). The teacher research movement: A
decade later. Educational Researcher, 28(7), 15–25.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.).
Hillsdale: Lawrence Erlbaum Associates.
Creswell, J. W. (2014). Educational research: Planning, conducting, and evaluat-
ing quantitative and qualitative research. Essex: Pearson.
Crookes, G. (1993). Action research for second language teachers: Going
beyond teacher research. Applied Linguistics, 14(2), 130–144.
Dikilitaş, K. (Ed.). (2012). Teacher-research studies at foreign language school:
Inquiries from teacher perspectives (Vol. 1). Ankara: Nobel.
Dikilitaş, K. (Ed.). (2013). Teacher-research studies at foreign language school:
Inquiries from teacher perspectives (Vol. 2). Ankara: Nobel.
Dikilitaş, K. (Ed.). (2014a). Professional development through teacher research.
İzmir: Gediz University Press.
Dikilitaş, K. (2014b). Introduction. In K. Dikilitaş (Ed.), Professional develop-
ment through teacher research. İzmir: Gediz University Press.
Dikilitaş, K. (2015). Teacher research for instructors. In S. Borg (Ed.), Professional
development for English language teachers: Perspectives from higher education in
Turkey (pp. 27–33). Ankara: British Council, Turkey.
Dikilitaş, K., & Mumford, S. E. (2016). Supporting the writing up of teacher
research: Peer and mentor roles. ELT Journal, 70(4), 371–381.
Dikilitaş, K., & Wyatt, M. (2017). Learning teacher-research-mentoring: Stories
from Turkey. Manuscript submitted for publication.
Dörnyei, Z. (2007). Research methods in applied linguistics. Oxford: Oxford
University Press.
Edge, J. (Ed.). (2001). Case studies in TESOL practice: Action research. Alexandria:
TESOL.
Edwards, E., & Burns, A. (2016). Language teacher action research: Achieving
sustainability. ELT Journal, 70(1), 6–15.
Farrell, T.  S. (2008). Critical incidents in ELT initial teacher training. ELT
Journal, 62(1), 3–10.
Freeman, D. (1996). Redefining the relationship between research and what
teachers know. In K. M. Bailey & D. Nunan (Eds.), Voices from the language
classroom (pp. 88–115). New York: Cambridge University Press.
Freeman, D. (1998). Doing teacher research. Boston: Heinle and Heinle.
296  References

Gibbs, G. (1988). Learning by doing: A guide to teaching and learning methods.

Oxford: Further Education Unit/Oxford Polytechnic.
Gieve, S., & Miller, I. K. (2006). What do we mean by ‘quality of classroom
life’? In Understanding the language classroom (pp.  18–46). New  York:
Palgrave Macmillan UK.
Glaser, B., & Strauss, A. (1967). The Discovery of grounded theory. New York:
Aldine.
Griffiths, C. (2003). Language learning strategy use and proficiency: The relation-
ship between patterns of reported language learning strategy (LLS) use by speakers
of other languages and proficiency with implications for the teaching/learning
situation. Ph.D. thesis, University of Auckland, New Zealand. http://hdl.
handle.net/2292/9
Griffiths, C. (2004). Language learning strategies: Theory and research.
Occasional Paper no. 1. Auckland, New Zealand, Centre for Research in
International Education. http://www.crie.org.nz
Griffiths, C. (Ed.). (2008). Lessons from good language learners. Cambridge:
Cambridge University Press.
Griffiths, C. (2015a). Dynamic strategy development and progress in language
learning. www.carolgriffiths.net
Griffiths, C. (2015b). What have we learnt from good language learners? ELT
Journal, 69(4), 425–433.
Griffiths, C. (2016). Strategies for developing English language writing skills.
Asian EFL (online).
Griffiths, C., & İnceçay, G. (2015). Styles and style-stretching: How are they
related to successful learning? Journal of Psycholinguistic Research, 45,
599–613.
Griffiths, C., & Özgür, B. (2013). Second language motivation. Procedia  –
Social and Behavioral Sciences, 70, 1109–1114.
Günbay, E. B., & Aydemir, G. (2015). Tackling speaking challenges faced by
low-level learners of English through consultation with students. In
K.  Dikilitaş, R.  Smith, & W.  Trotman (Eds.), Teacher-researchers in action
(pp. 97–108). IATEFL: Faversham.
Güzel, M. (2015). Hey, teacher, leave those kids alone! Beliefs of students and
instructors about learner autonomy. In K. Dikilitaş, R. Smith, & W. Trotman
(Eds.), Teacher-researchers in action (pp. 189–206). IATEFL: Faversham.
Halai, A. (2006). Mentoring in-service teachers: Issues of role diversity. Teaching
and Teacher Education, 22(6), 700–710.
  References 
   297

Hanks, J.  (2009). Inclusivity and collegiality in exploratory practice. In

Researching language teaching and learning: An integration of practice and the-
ory (pp. 33–55). New York: Peter Lang.
Hanks, J. (2014). ‘Education is not just teaching’: Learner thoughts on explor-
atory practice. ELT Journal, ccu063.
Hanks, J.  (2015). Language teachers making sense of exploratory practice.
Language Teaching Research. doi:10.1177/1362168814567805.
Hendricks, C. (2009). Improving schools through action research: A comprehensive
guide for educators (2nd ed.). Boston: Allyn & Bacon.
Hobson, A. J., & Sharp, C. (2005). Head to head: A systematic review of the
research evidence on mentoring new head teachers. School Leadership &
Management, 25(1), 25–42.
Horwitz, E. (1987). Surveying student beliefs about language learning. In
A.  Wenden & J.  Rubin (Eds.), Learner strategies in language learning
(pp. 119–132). London: Prentice Hall.
Horwitz, E., Horwitz, M., & Cope, J. (1986). Foreign language classroom anxi-
ety. The Modern Language Journal, 70(2), 125–132.
Hu, H., & Nassaji, H. (2014). Lexical inferencing strategies: The case of success-
ful versus less successful inferencers. System, 45, 27–38.
Kordia, S. (2015). From TEFL to ELF-aware pedagogy: Lessons learned from
an action-research project in Greece. In K. Dikilitaş, R. Smith, & W. Trotman
(Eds.), Teacher-researchers in action (pp. 235–262). IATEFL: Faversham.
Labov, W. (1972). Sociolinguistic patterns. Philadelphia: University of
Pennsylvania.
Landsberger, H. (1958). Hawthorne revisited. New York: Ithaca.
Lewin, K. (1946). Action research and minority problems. Journal of Social
Issues, 2(4), 34–46.
Lewis, K. A. (2009). Adult learners’ perceptions of the incorporation of their
L1  in foreign language teaching and learning. Applied Linguistics, 30(2),
216–235.
Lewis, C., Perry, R., Hurd, J., & O’Connell, M. P. (2006). Lesson study comes
of age in North America. Phi Delta Kappan, 88(4), 273.
Lieberman, A. (1995). Practices that support teacher development: Transforming
conceptions of professional learning. Phi Delta Kappan, 76, 591–596.
Likert, R. (1932). A technique for the measurement of attitudes. Archives of
Psychology, 140, 1–55.
Lin, M., Li, J., Hung, P. and Huang (2014). “Blogging a journal: Changing
students’ writing skills and perceptions”. ELT Journal, 68/4, 422-431.
298  References

Little, D. (1995). Learning as dialogue: The dependence of learner autonomy on

teacher autonomy. System, 23(2), 175–181.
Livingston, K., & Shiach, L. (2013). Teaching Scotland’s future. Mentoring Pilot
Partnership Project, Final report.
Loughran, J.  J. (2002). Effective reflective practice in search of meaning in
learning about teaching. Journal of Teacher Education, 53(1), 33–43.
Lynch, B.  K. (1996). Language program evaluation: Theory and practice.
Cambridge/New York: Cambridge University Press.
MacLean, M. S., & Mohr, M. M. (1999). Teacher-researchers at work. National
Writing Project, University of California, 2105 Bancroft,# 1042, Berkeley,
CA, 94720-1042.
Malderez, A., & Bodóczky, C. (1999). Mentor courses: A resource book for trainer
trainers. Cambridge: Cambridge University Press.
McDonough, J., & McDonough, S. (1997). Research methods for language teach-
ers. London: Hodder Arnold.
McGrath, I. (2000). Teacher autonomy. In B.  Sinclair et  al. (Eds.), Learner
autonomy, teacher autonomy: Future directions. Harlow: Pearson Education.
Mitchell, S. N., Reilly, R. C., & Logue, M. E. (2009). Benefits of collaborative
action research for the beginning teacher. Teaching and Teacher Education,
25(2), 344–349.
Myhill, D. (2002). Bad boys and good girls? Patterns of interaction and response
in whole class teaching. British Educational Research Journal, 28(3), 339–352.
Nunan, D. (1989). Designing tasks for the communicative classroom. Cambridge/
New York: Cambridge University Press.
Nunan, D. (1992). Research methods in language learning. Cambridge: Cambridge
University Press.
Orland-Barak, L., & Rachamim, M. (2009). Simultaneous reflections by video
in a second-order action research-mentoring model: Lessons for the mentor
and the mentee. Reflective Practice, 10(5), 601–613.
Oxford, R. (1990). Language learning strategies: What every teacher should know.
New York: Newbury House.
Patton, M. Q. (1987). How to use qualitative methods in evaluation. Newbury
Park: Sage.
Polat, B. (2013). Language experience interviews: What can they tell us about
individual differences? System, 41(1), 70–83.
Porte, G. (2012). Replication research in applied linguistics. Cambridge:
Cambridge University Press.
  References 
   299

Raelin, J. A. (2002). I don’t have time to think! versus the art of reflective prac-
tice. Reflections, 4(1), 66–79.
Rebolledo, P., Smith, R., & Bullock, D. (Eds.). (2016). Champion teachers:
Stories of exploratory action research. London: The British Council.
isbn:9780863558078.
Reid, J. (1987). The learning style preferences of ESL students. TESOL Quarterly,
21(1), 87–111. doi:10.2307/3586356.
Richards, J. C., & Burns, A. (2012). The Cambridge guide to pedagogy and prac-
tice in second language teaching. Cambridge/New York: Cambridge University
Press.
Richards, J. C., & Farrell, T. S. C. (2005). Professional development for language
teachers: Strategies for teacher learning. Ernst Klett Sprachen. Cambridge/New
York: Cambridge University Press.
Richards, J., & Farrell, T. (2011). Practice teaching: A reflective approach.
Cambridge: Cambridge University Press.
Schön, D. (1983). The reflective practitioner. New York: Basic Books.
Schön, D. (1988). Educating the reflective practitioner. San Francisco:
Jossey-Bass.
Seidman, I. (2006). Interviewing as qualitative research: A guide for researchers in
education and the social sciences (3rd ed.). New York: Teachers College Press.
Smith, R. C. (2000). Starting with ourselves: Teacher-learner autonomy in lan-
guage learning. In B. Sinclair, I. McGrath, & T. Lamb (Eds.), Learner auton-
omy, teacher autonomy: Future directions (pp. 89–99). London: Longman.
Smith, R. C. (2003). Teacher education for teacher-learner autonomy. In J.
Gollin, G. Ferguson, & H. Trappes-Lomax (Eds.), Symposium for language
teacher educators: Papers from three IALS symposia (CD-ROM). Edinburgh:
IALS, University of Edinburgh. http://homepages.warwick.ac.uk/~elsdr/
Teacher_autonomy.pdf
Smith, R. C. (2015). Exploratory action research as workplan: Why, what and
where from? In K.  Dikilitaş, R.  Smith, & W.  Trotman (Eds.), Teacher-­
researchers in action (pp. 235–262). IATEFL: Faversham.
Smith, R., Connolly, T., & Rebolledo, P. (2014). Teacher research as continuing
professional development: A project with Chilean secondary school teachers.
In D. Hayes (Ed.), Innovations in the continuing professional development of
English language teachers (pp. 111–129). London: British Council.
Soruç, A., & Griffiths, C. (2015). Identity and the spoken grammar dilemma.
System, 50, 32–42.
Stenhouse, L. (1975). An introduction to curriculum research and development.
London: Heinemann.
300  References

Strauss, A., & Corbin, J. (1998). Basics of qualitative research: Techniques and
procedures for developing grounded theory (2nd ed.). Thousand Oaks: Sage.
Tammenga-Helmantel, M., Arends, E., & Canrinus, E. (2014). The effective-
ness of deductive, inductive, implicit and incidental grammatical instruction
in second language classrooms. System, 45, 198–210.
Tseng, W., Dornyei, Z., & Schmitt, N. (2006). A new approach to assessing
strategic learning: The case of self-regulation in vocabulary acquisition.
Applied Linguistics, 27(1), 78–102.
Ullman, R., & Geva, E. (1981). The target language observation scheme (TALOS):
Handbook. Toronto: Modern Language Center/Ontario Institute for Studies
in Education.
Velde, N. (2015). Team teaching for teacher training. In K. Dikilitaş, R. Smith,
& W. Trotman (Eds.), Teacher-researchers in action (pp. 189–206). IATEFL:
Faversham.
Vieira, F., Barbosa, I., Paiva, M., & Fernandes, I. S. (2008). Teacher education
towards teacher (and learner) autonomy. In Learner and teacher autonomy:
Concepts, realities, and responses (Vol. 1, p. 217). Amsterdam: John Benjamins
Pub. Co.
Wallace, M. J. (1991). Training foreign language teachers: A reflective approach.
Cambridge/New York: Cambridge University Press.
Wang, Q., & Zhang, H. (2014). Promoting teacher autonomy through univer-
sity–school collaborative action research. Language Teaching Research, 18(2),
222–241.
Wilson, D. Practical meta-analysis effect size calculator. https://www.campbell-
collaboration.org/escalc/html/EffectSizeCalculator-R5.php
Wyatt, M., & Arnold, E. (2012). Video-stimulated recall for mentoring in
Omani schools. International Journal of Mentoring and Coaching in Education,
1(3), 218–234.
Wyatt, M., & Dikilitaş, K. (2015). English language teachers becoming more
efficacious through research engagement at their Turkish university.
Educational Action Research, 1–21. DOI: 10.1080/09650792.2015.1076731.
Wyatt, M., & Márquez, C.  P. (2015). Helping first-year undergraduates
engage in language research. Language Teaching Research. doi:10.1177/
1362168814562013.
Yanati, K. http://yatani.jp/teaching/doku.php?id=hcistats:mannwhitney#effect_size
Yu, X. (2009). A formal criterion for identifying lexical phrases: Implications
from a classroom experiment. System, 37(4), 689–699.
Yuan, R., & Lee, I. (2015). Action research facilitated by university–school col-
laboration. ELT Journal, 69(1), 1–10.
 Index

A conference presentations,
abstract, 172, 184–5, 203, 224, 218–21
230–1, 273 correlations
action research, 1–43, 46, 54, 58–60, Pearson’s, 130, 228
63, 84, 86–7, 91–100, 102, Spearman’s, 130, 144, 228
104–5, 109–10, 113–14, 116, criticality, 166–9
127, 129, 131–2, 155–9, cross sectional, 64–5, 227
161–3, 165–7, 171, 191,
217–35, 250, 255, 258–61,
263, 265–8, 273 D
American Psychological Association debriefing, 154–5
(APA), 151, 230, 236 deduction, 66–7, 151–2, 249
ANOVA/MANOVA, 67, 130, 137, deductive analysis, 159
149, 151, 228 descriptive statistics, 139–41, 149
autonomy, 2, 29, 33, 35–9, 83, 100, differences
165–6, 243, 264–70 Kruskal–Wallis, 67, 130–1,
149
Mann–Whitney, 67, 130–1,
C 146–7, 207, 228, 275–6
case study, 81, 84, 171, 190, 252–6 t-tests, 67, 130, 137, 145, 178–9,
collaborative action research, 12–13 228

© The Author(s) 2017 301

K. Dikilitaş, C. Griffiths, Developing Language Teacher Autonomy
through Action Research, DOI 10.1007/978-3-319-50739-2
302  Index

dissemination, 7, 9, 16–17, 23, 41, I

221–2, 233 induction, 62, 66–7, 152, 161, 242,
documentation, 85, 192 249
inductive analysis, 20, 151, 160
inferential statistics, 142
E in-house presentations, 218
effect size, 85, 132, 149–51 inquiry, 7, 23, 29, 31, 35–8, 100,
ethics 264
absence of threat, 110 based learning, 35, 100
anonymity, 109 process-oriented, 8
confidentiality, 109–10 systematic, 22, 38
informed consent, 108 inter-rater reliability, 18, 154
professional distance, 110 interviews
storage of data, 110 semi-structured, 4, 78, 123, 211
voluntary participation, 108–9 structured, 4, 78, 123, 211
ethnography, 63, 82–3, 228 unstructured, 4, 78, 123
experimental, 24, 46, 65–7, 80, introspection, 79–80
174–7, 179–80, 183, 191,
227–8, 244
exploratory action research, 31 K
exploratory practice, 15, 24–30, 34, keywords, 224, 231, 273
49, 97, 249–52

L
F lesson study, 21, 33–4, 255
factor analysis, 132, 134–7 Likert-type questionnaire, 63, 67–8,
feasibility, 39–40, 56, 89, 263 142, 151, 155, 228
literature review, 23, 174–5, 186–90,
204–5, 222, 224–5, 231,
G 273–4
grounded theory, 81, 83–4, 151, longitudinal, 64–5, 81, 191, 227
191
open, axial, selective coding, 84
M
mean, 85, 132, 137, 139
H median, 85, 132, 137, 139–42, 146,
Hawthorne effect, 75, 98 167, 207–8, 210–11, 275–6,
hypothesis/es, 8, 32, 45–59 278
  Index 
   303

member checking, 82, 154, 156 163–5, 167, 174–6, 182–3,

mixed methods, 84, 86, 151, 171, 190, 192, 206, 226–8, 231–2,
273–80 274–5, 278
mode, 85, 132, 137, 139, 141–2 phenomenology, 83
piloting, 85–6, 115, 263
population, 7, 18, 81, 95, 142, 165,
N 190
narrative/s, 81–2, 84, 235, potentially exploitable pedagogic
237–61 activities (PEPAs), 26, 29, 34,
actual studies, 238–48 97, 251
meta-narratives, 256–61 problematize, 3, 6, 21, 32, 36, 45,
teacher’s personal narratives, 66, 93, 95, 99, 218, 269
248–56 professional development, 2, 7–8,
nominal, 131, 142, 145 10–11, 15, 17, 22, 24, 33–5,
nonparametric, 63, 67, 130–1, 38–41, 43, 50, 92, 161, 166,
137–8, 142, 145, 149, 151, 249, 264, 268, 270
207, 228, 275–6 purposive sampling, 212
normal distribution, 179, 207
null hypothesis, 46
Q
qualitative, 5, 20, 46, 48, 52, 59,
O 61–4, 79, 81–2, 84, 112,
observation, 4, 28, 33–4, 46, 48, 63, 123–4, 151–2, 154, 156, 168,
73–7, 81, 84, 98, 113, 124–7, 191, 194, 227–9, 263, 267,
152–4, 190, 227–8, 238, 245, 273, 275–6, 280
249, 251, 261 quality of research, 12, 17–19, 253
observer paradox, 74 quantitative, 5, 20, 48, 52, 59, 61–4,
online presentation, 221 73, 81, 84, 129–31, 151, 153,
ordinal, 63, 67, 130–1, 138–9, 142, 155–6, 191, 227–8, 257–8,
155 263, 273
quasi-experimental, 65–6, 84,
171–83, 227–8
P questionnaire, 4, 17, 46, 56, 59,
parametric, 130, 137, 149, 151, 207, 63–5, 67–73, 83–5, 87, 103,
228 111, 113–23, 127, 130, 132,
participants, 1–2, 7, 15, 53, 63, 135–6, 142, 150–1, 155–6,
65–72, 74, 77–9, 82, 85, 168–9, 171, 191–4, 203–5,
89–105, 108–10, 112, 131, 227–8, 230–1, 238–9, 257–9,
134, 140, 151–2, 154, 159, 261, 273, 275–6, 280
304  Index

R T
reflective practice, 21, 31–3, 256–7 teacher autonomy, 2, 35–9, 264–8
regression, 130 teacher development, 2, 31–2, 35,
reliability, 17–18, 70, 72, 79–81, 96, 37, 191, 248
109, 115, 117, 132–5, 137, teacher research, 17, 21–4, 36, 38,
154, 156, 207, 251 45, 47, 53, 91, 99–102, 126,
replication, 171–2, 237 218, 230, 232, 237, 248–56
reports, 42, 50–1, 65, 80–2, 177 think aloud, 63, 79–80, 84, 112, 211
research question guidelines, 55–6 triangulation, 81, 84–5, 113, 211,
263
t-test, 67, 130, 137, 145, 178–9, 228
S
sample, 47, 56, 62, 78, 95, 124–5,
133–4, 136–42, 144–9, 165, V
178–9, 211–12, 220 validity, 17–18, 72, 81–2, 96,
setting/context, 89–105, 179, 192, 189–90, 211, 251, 253, 267–8
205–6
stages of action research, 19–21, 263
standard deviation, 139 W
sustainability, 11, 26, 40–3, 91, written presentations, 223–30
268–9