Grounded theory (GT) approach to accounting studies

Overview of principles, assumptions, and methods

There have been many calls for developing theories of accounting from qualitative case

studies. These have occurred particularly in the area of management accounting (Humphrey

and Scapens 1996, Llewellyn, 2003, Ahrens and Chapman 2006, Elharidy, Nicholson and

Scapens, 2008). GT provides a coherent set of methods to develop such theories and is

particularly appropriate when studying accounting in its social context. Indeed, it has become

a popular qualitative research methodology in many social disciplines (see below). However

it has received relatively little attention form accounting researchers. The purpose of this

chapter, therefore, is to provide an introduction to GT. The chapter commences with an

outline of the various approaches taken to GT by researchers in other disciplines. This is

followed by a practical example of the use of GT in an accounting research project to

illustrate the methods used. Next a brief review of accounting GT research to date is

presented followed by a discussion of some issues and debates which are emerging. My own

personal experiences of GT in the field are presented to guide potential researchers. The

chapter concludes with an outline of the possible future of GT in accounting research

This chapter would ideally commence with a precise and succinct definition of GT.

Unfortunately it is not so straightforward and there seem to be as many definitions as

researchers. GT also seems to engender a great deal of misunderstanding. This leads the

researcher new to GT facing a number of questions regarding GT such as is it qualitative or

quantitative, is it inductive or deductive, is it interpretive or functionalist, is it subjective or

objective, is it methodology or method and what is its relationship to prior theory. I hope to
address these and other questions in the following chapter to enable the researcher to make

their own mind up and hopefully to adopt its use.

Approaches to GT

In this section the core aspects of GT are introduced together with brief outlines of its

consequent developments and alternative approaches.

Traditional GT of Strauss and Corbin

Historically, GT emerged from the Chicago School of Sociology and the development of

symbolic interactionism during the period 1920 - 1950 (Kendall, 1999). Both the school and

the symbolic interactionism theorists heavily criticised the functionalist paradigm that

dominated the sociological domain at the end of the 19 th century till middle of the 20th century

and sought alternative approaches. Two sociologists named Barney Glaser and Anselm

Strauss first coined the term GT and formally introduced its concepts in their well-known

book, The Discovery of Grounded Theory, published in 1967. They addressed the question of

“how the discovery of theory from data – systematically obtained and analysed in social

research – can be furthered” (Glaser and Strauss, 1967, p. 1). Although originating from the

field of sociology, GT has since been used in areas as diverse as psychology, anthropology,

education, social work, nursing, accounting and management, to name just a few.

There have been many attempts to summarise GT (Charmaz 2006, Gurd 2008, Elharidy,

Nicholson and Scapens, (2008). Charmaz (2006, p. 5 – 6) suggests the key components of

GT practices, as advocated by its founders and incorporating most elements identified by

other researchers, are as follows.

(a) Simultaneous and iterative involvement in data collection and analysis. Corbin and

Strauss (1990) make this the first of their “canons and procedures”.

(b) Constructing analytic codes and categories from data, not from preconceived logically

deduced hypotheses.

(c) Using the constant comparative method. This involves making comparisons during

each stage of the analysis..

(d) Advancing theory development during each step of data collection and analysis. The

aim of GT is not to “test” the emerging propositions, but to be open to new avenues and to be

prepared for “surprises” in the field (Elharidy Brian Nicholson Robert W. Scapens, 2008)

(e) Memo-writing to elaborate categories, specify their properties, define relationships

between categories, and identify gaps in the categories;

(f) Sampling aimed towards theory construction, not for population representativeness

(theoretical sampling).

(g) Conducting the literature review after developing an independent analysis. However,

this is a very contentious area and many GT researchers would insist that at least a general

knowledge and understanding of prior research in the substantive area is advisable (see

below).

What distinguishes GT from other research methods is the systematic process for data

collection and analysis. This process has been developed over four editions of the core text

entitled The Basics of Qualitative Research, comprising Strauss and Corbin (1990,1998) and

Corbin and Strauss (2008,2015), all of which are essential reading for the new researcher.

These texts constitute traditional GT and explain the three (main) processes of coding: open,

axial and selective.

Open coding is the initial and provisional production of concepts that ‘opens up’ the data. It

is the opening attempt to fracture the data and allow one to identify some categories, their

properties and dimensions (Strauss and Corbin, 1990), and to get the researcher less

immersed in the literal dimension of data, but more immersed in concepts and their

relationships (Strauss, 1987). Open coding often stimulates the researcher’s though processes

which can be captured by use of theoretical memos. Memos are ‘a running record of insights,

hunches, hypotheses, discussions about implications of codes, additional thoughts, etc.’

(Strauss, 1987, pp. 30-31), and are an essential part of theory generation. They document the

continuing internal dialogue that ensues between the researcher, laden with experiential data,

and the generated data, and ceases only when the research terminates. The theoretical depth

of a memo varies with the phases of the research and as work progresses to higher levels

memos embody the cumulative results of earlier efforts and get more conceptual, focusing on

emerging categories, relationships between categories and theory.

Axial coding is a task that builds and consolidates open coding by intense analysis of

categories along larger relationships that transcend beyond the early dimensions and

properties of open codes. The focus is on linking and amalgamating open categories to

establish broader, more abstract categories. The product of axial coding is a rich mesh of

main categories sufficiently linked to enable some aggregation, patterning and generation of

potential core categories that on further analysis could uphold a theory.

Selective coding is axial coding executed at a higher and even more abstract level (Strauss

and Corbin, 1990; 1998). A concerted effort is made to systematically bring out core

categories (Strauss, 1987) by concentrating on prominent categories and observing and

strengthening relationships between these and other subservient categories. Selective coding

also involves filling in empirical indicators in gaps in relationships, a process that is guided

by theoretical sampling. The core categories form the backbone to the theory while their
relationships with other subservient categories provides patterns, fills in the richness and

conceptual density necessary for good theory.

Axial and selective coding may be aided by use of the paradigm model (Strauss and Corbin,

1990; 1998). The paradigm model seeks to link categories in terms of answers to questions

'who, why, where, how, and with what consequences' (Strauss and Corbin, 1998, p. 127) and

facilitates the integration of elements of structure with those of process. Strauss and Corbin

(1998 p. 128) describe the basic components of the paradigm model as, ‘there are conditions,

a conceptual way of grouping answers to the questions why, where, how come, and when.

Together, these form the structure, or set of circumstances or situations, in which phenomena

are embedded. There are actions/interactions, which are strategic or routine responses made

by individuals or groups to issues, problems, happenings, or events that arise under those

conditions. Actions/interactions are represented by the questions by whom and how. There

are consequences, which are outcomes of actions/interactions. Consequences are represented

by questions as to what happens as a result of those actions/interactions or failure of persons

or groups to respond to situations by actions/interactions, which constitutes an important

finding in and of itself. ’

Consequent Developments and Alternative Approaches

The systematic analytical process is one of the greatest strengths of GT. It provides a clear

audit trail from the raw data to the GT itself. However the rigidity of the process is also a

source of disagreement between GT researchers and beyond. This alongside other

refinements has led to the development of a number of different approaches to GT.

The first results from Glaser and Strauss having differing views on what they consider GT to

be, revolving around five major issues. Firstly, Strauss regarded the GT as the research
method (technique) which is used to execute the qualitative study. Strauss’s definition of GT

excludes the possibility of executing the quantitative study by using the GT approach. On the

other hand, Glaser viewed GT as a general methodology which could be applied in both

qualitative and quantitative study. Secondly, Strauss favoured the researcher having prior

knowledge of the phenomena or issues to be studied (Strauss & Corbin, 1990; Strauss &

Corbin, 1998). In contrast, Glaser recommended that the researcher should select an

organisation and allow the issues or phenomena to emerge in the course of the research

process (Glaser, 1992). Thirdly, Glaser preferred a general approach to data analysis which

allows the issue of interest to emerge in the field (Glaser, 1992). On the other hand, Strauss

favoured a more structured analytical method (Strauss & Corbin, 1990; Strauss & Corbin,

1998), which was regarded by Glaser as “forcing” issues to emerge instead of emerging

themselves in the research process. Finally, Strauss regarded formal GT as elevating the

“concept of the study up to a more abstract level where it can have broader applicability but

at the same time remain grounded in data” (Corbin & Strauss, 2008, p. 102). In contrast,

Glaser described formal GT as the general implications of the core categories which emerged

from a substantive theory. In practical terms, Strauss and Corbin’s approach has proven to be

more popular with researchers.

As Glaser and Strauss were developing GT at the University of California, SanFrancisco

(UCSF), Leonard Schatzman (1991) was involved in the development of another approach to

GT, called dimensional analysis. Although procedures used in dimensional analysis are

consistent with those of GT method, dimensional analysis has its own epistemology and

unique set of operations. Strauss and Corbin’s (1990) multiple coding procedures introduced

a level of complexity into the analytic process that could be claimed to divert the researcher

from generating theory directly from data and risk the reduction of theoretical sensitivity

(Glaser, 1992; Robrecht, 1995). Schatzman addressed the complexity by embedding

dimensional analysis in symbolic interactionism. Dimensional analysis is not as prevalent in

the literature as traditional GT. However, the concept of properties and dimensions were

incorporated in the first edition of Strauss and Corbin although they were substantially toned

down in later editions.

More recently, Charmaz (2006, 2014) developed GT from what she claims are the positivistic

roots of Glaser, Strauss and Corbin within an interpretive/social constructivist approach. She

suggests that GT strategies such as coding, memo writing, theoretical sampling and

comparative methods are ‘transportable across epistemological and ontological gulfs,

although which assumptions researchers bring to these strategies and how they use them

presuppose epistemological and ontological stances’ (Charmaz 2014 p 12.). Perhaps the

principal assumption the social constructivist brings to GT is that neither data nor theories are

discovered but are constructed by the researcher and research participant (Allen 2010).

Moreover, Charmaz emphasises an interpretive approach whereby participants’ meanings,

experiential views – and researchers’ finished grounded theories – are constructions of

reality. In terms of how the strategies are used, social constructivist places greater emphasis

on the purpose and practice of interviewing with a view to eliciting meaning and

acknowledging the importance of language. Charmaz also takes a far less prescriptive

approach to data analysis with a simpler approach to coding and no use for detailed analysis

of code properties or the paradigm model. Indeed, ‘constructivist GT highlights the flexibility

of the method and resists the mechanical operation of it’, (Charmaz 2014, p 13). This also

meets some of the criticisms of GT relating to the over rigorous, positivistic approach to

analysis. Charmaz’s approach to GT is proving to be very popular particularly with social

constructivist and interpretivist researchers and with accounting researchers.

The most recent development of GT has been made by Adele Clarke who has attempted to

bring a post modernist perspective to the approach (Clarke 2003,2005). Clarke (2003, p559)
notes that her approaches to GT are congruent with Charmaz’s social constructivism and that

‘Charmaz emphasizes that a focus on meaning making furthers interpretive, constructivist,

and, I would add, relativist/perspectival understandings’. She identifies traditional GT as

concerned with modernism which emphasized universality, generalization, simplification,

permanence, stability, wholeness, rationality, regularity, homogeneity, and sufficiency.

However, postmodernism has shifted emphases to localities, partialities, positionalities,

complications, tenuousness, instabilities, irregularities, contradictions, heterogeneities,

situatedness, and fragmentation-complexities. In order to allow GT to encompass

postmodernism its root metaphor of social process/action needs to be supplemented with an

ecological root metaphor of social worlds/arenas/negotiations. This allows situational

analyses at the mesolevel, new social organizational/institutional and discursive sitings, as

well as individual-level analyses (Clarke 2003). In practical terms this is achieved by

supplementing the traditional GT analysis with situational mapping.

To summarise traditional GT as developed by Strauss and Corbin has certainly proved the

most popular approach to date and has come to stand for what GT means to most researchers.

However, Charmaz and Clarke may well generate more interest as time goes on. Indeed in

their latest edition Corbin and Strauss (2015) have at least attempted to respond to some of

the issues raised by Charmaz and Clarke.

A practical example

It may be helpful to illustrate the coding methods and theory development using grounded

theory by way of an actual research project. This project sought to understand accounting

processes and reporting practices in non-governmental organisations (NGOs) and the

conditions that sustained those processes and practices. Fieldwork research was undertaken in
two phases over a period of fifteen months in three in NGOs in Tanzania. The principal

methodology for the inquiry was that of Strauss and Corbin (1990, 1998).

The main source of data was from 31 tape-recorded interviews with stakeholders concerned

with the financial management of the NGO’s. Data was analysed using Strauss and Corbin’s

recommended procedures. Open coding was undertaken initially and was principally

undertaken in a sentence or paragraph mode. Open coding normally results in large numbers

of concepts and by the end of the first fieldwork, for example, 78 concepts had been

identified and labelled from research data. These were subsequently reduced to twenty

categories by grouping related concepts. Axial coding was undertaken next which focused on

linking categories or phenomena together and developing the relationships with other

categories. Fig. 1 depicts in a code matrix the output of this coding process which shows the

relationships between the initial twenty categories from open coding and the main categories

emerging from the axial coding.

<Fig 1 here>

Finally, selective coding was undertaken to establish the central phenomenon and how this

related to other main categories. Use was made of a simplified version of Strauss and

Corbin’s paradigm model to compose and present the resultant grounded theory. This

comprised the core category or central phenomenon, the action/interaction strategies, the

conditioning context and finally the consequences. The paradigm model of the final grounded

theory is illustrated in Figure 2.

<Fig 2 here>
The emergent grounded theory was anchored around the central phenomenon - the basic

process of navigating legitimacy. The main story from the data was how, and the extent to

which, NGOs succeeded in accessing resources from donors and the modes by which these

organisations justified resource utilisation to a spectrum of stakeholders. The organisations

solely depended on donated resources for their existence. Consequently, resource attainment

and utilisation were critical to the way organisations attained, lost, maintained or enhanced

legitimacy. The nature of their responses to, and interactions with, stakeholders can be

summarised as a process of navigating legitimacy.

Two macro-conditioning contexts were identified as facilitating the phenomenon of

navigating legitimacy. These were firstly; increased availability of easily accessible donor

funds, combined with secondly; a weak and outdated regulatory regime. It was the

combination of these two conditions that appeared to explain the phenomenon of

'mushrooming of NGOs' - the high rate of unregulated growth in NGOs in the past ten years.

Two micro conditioning contexts were also identified. These were the internal organisational

situation; and accountability profiles of individual organisations. Taking account of these

micro conditioning contexts organisations appeared to seek either superior accountability

[high legitimacy threshold] or inferior accountability [low legitimacy threshold].

Organisations seeking a high legitimacy threshold seemed to engage the whole spectrum of

stakeholders while those seeking a low legitimacy threshold seemed to engage only the most

critical stakeholders who were the Registrar of Societies and donor institutions. These

organisations appeared to keep a low public profile, maintained the minimum house-keeping
requirements and met donor accounting and reporting requirements to ensure a continued

flow of funds.

A number of strategies and tactics that organisations employed in navigating legitimacy were

identified; building credibility; and bargaining for change. It was shown that during the phase

of building credibility organisations seemed to initially seek to establish a good track record

with key stakeholders, especially funding agencies. It was followed by a stage where

organisations sought to consolidate long-term relationships with key stakeholders such as

donors and government officials in sensitive positions. Finally having established a good

track record and appropriate support among key stakeholders organisations attempted to

lessen, through a bargaining strategy, the burden of multiple accounting and financial

reporting.

Accounting emerged as predominant in the whole process of navigating legitimacy, which

suggested that despite its acknowledged inconsequential role in internal decision processes,

accounting was a crucial legitimating device. The identification of organisations with sound

management practices such as accounting portrayed them as competent entities. The

utilisation of expert professionals such as management consultants and auditors was also

shown to have a similar effect of legitimating an entity. Over time, the profile of accounting

functions within organisations seemed to have substantially increased. Evidently, there were

on going endeavours to incorporate 'sound' management practices. Nevertheless,

organisational actors perceived 'managerialism' as a threat to the continued survival of the

activist orientations established and nurtured in organisations.

More details on this project and a discussion of how the final theory related to prior research

on accounting and legitimacy can be found in Goddard and Assad (2006).

Brief review of accounting GT research to date

GT has been used extensively in social studies, nursing and education research over the last

25 years. More recently there has been substantial use of GT in management and

organisational research. For instance in the last ten years there have been well over 100

papers which report the use of GT published in one of the world’s leading management

journals - the Academy of Management Journal. Despite exhortations and supportive papers

for the use of GT in accounting research (Parker and Roffey 1995, Elharidy, Nicholson and

Scapens, 2008, Joannides and Berland, 2008) there have been relatively few such studies.

Gurd (2008) found 24 GT studies published in leading accounting journals up to 2006. A

brief review of these journals up to 2015 adds less than a dozen further studies. The vast

majority of GT accounting studies have been undertaken in the broad area of management

accounting and control, with only a few in the areas of financial reporting and audit.

Covaleski and Dirsmith, (1983,1984) were probably the first accounting researchers to use

GT but it was not until the late 1990’s that the approach became more established. Parker and

Roffey’s (1997) seminal paper certainly inspired a number of researchers to explore the use

of GT. The contributions of GT to knowledge and understanding of accounting is still limited

due to the paucity of publications but nonetheless some clear themes are emerging. Joannides

and Berland (2008) suggest that GT has the ability to produce theoretical, empirical (and

methodological) contributions. They evaluated the sample of papers selected by Gurd (2008)

to assess these contributions. In terms of theoretical contributions they suggest that the

discovery of formal theories (those with a higher degree of generalisability) is rare; rather,

grounded theorists generate substantive theories or confirmatory theories (those confined to

the context in which they were developed). Formal theories are apparent when prior research

in a substantive area has omitted theoretical findings of a GT study but subsequent research
references them extensively. They cite Ahrens (1996) and Wickramasinghe et al. (2004) as

examples of such formal theoretical development in the areas of accounting and

accountability and accounting in developing countries respectively. Most accounting GT

studies however develop substantive theories and call for further investigations. Examples

include Lightbody (2000), Parker (2001) and Parker (2002) in the area of accounting in

religious organisations, Solomon and Solomon (2006) in the area of private ethical, social

and environmental disclosures, Efferin and Hopper (2007) in control of multi-ethnic

organisations and Goddard and Assad (2006) in accounting and NGOs. To these might be

added Beattie, Fearnley and Brandt (2004) in auditor/client negotiations, Goddard

(2004,2005) in accounting in UK local government and Goddard and Assad (2010), Goddard

and Issa (2015), Goddard and Malagila (2015), Goddard and Mkasiwa (2015) and Goddard,

Assad, Issa, Malagila and Mkasiwa (2015) in accounting in emerging economies. Consistent

with Strauss and Corbin (1998), these cannot be considered as formal theories yet; until

further works enrich or confirm them, they are regarded as substantive theories.

GT can also make use of and contributions to prior theory. The extent and way in which GT

should engage with prior theory is a matter for debate (see below) however several studies

have made extensive use of other ‘schools of thought’ to provide perspectives on their

research (Locke, 2001). For instance Coopey, Keegan and Emler (1998) used structuration

theory as an orientating framework alongside GT methods of analysis in their study of

innovation in organisations. Covaleski et al. (1998) used a Foucauldian framework in their

GT study of mentoring in accounting firms. Goddard et al (2015) develop their GT theory of

public sector accounting in a post-colonial context, alongside post-colonial and institutional

theory.

Joannides and Berland (2008) also note that ‘the most obvious empirical contribution of

grounded theorists consists of rich description’. They do suggest many of the papers are
evasive or allusive regarding empirics but this is most likely due to the requirements of

publication as rich description is the very core of GT. Indeed GT research which does not

provide rich empirical understanding of a substantive area cannot claim to be GT at all.

Elharidy, Nicholson and Scapens (2008) suggest that GT is particularly useful in undertaking

interpretive management accounting research, ‘it clearly fits the essential aim of interpretive

management accounting research, which is to produce theories of management accounting

research’ (p149). They do however warn against over-prescriptive adherence to GT methods

and procedures and provide some general guidelines for its use.

Personal experiences of GT in the field

My GT research

I have undertaken many GT accounting research studies over the last fifteen years. These

have been in my main areas of interest, management accounting, accounting and

accountability and performance management in the public and not for profit sector. They

have been undertaken in various settings including local government in the UK and

Indonesia, UK universities and health sector, NGOs in the UK and Africa, religious

organisations in the UK and Asia, public sector of Brunei and even strategic management

accounting in a German company. In most cases these have been undertaken in partnership

with doctoral students and other researchers. A range of approaches have been used from a

fairly rigorous application of Strauss and Corbin’s methods and procedures to looser

approaches more akin to Glaser and Charmaz. In almost all cases the research has been

published in good journals (Goddard 2004,2005, Goddard and Tillman 2008, Goddard and

Assad 2010, Goddard and Issa 2015, Goddard and Malagila 2015, Goddard and Mkasiwa
2015 and Goddard, Assad, Issa, Malagila and Mkasiwa 2015) though after much effort and

compromise (see below).

Most of the studies have resulted in substantive theories but as the research has developed

more deliberate attempts have been made to develop more formal theories. An example of the

latter is a series of research projects in public sector financial management in Tanzania. Three

of these studies were concerned with the effects of changes in accounting emanating from

new public management reforms in central government, local government and non-

governmental organisations (NGOs) in Tanzania. Within each setting a substantive theory

was developed. These were then compared, contrasted and considered in the light of post-

colonial and institutional theory to develop a broader formal theory. Another example

comprised four studies of accounting in NGOs in the UK, Tanzania and Zimbabwe. Again,

four substantive theories were separately developed and then compared, contrasted and

considered in the light of the social theories of Bourdieu to develop a more formal theory.

Pitfalls and advice

Over the years my fellow researchers, students and I have encountered a number of potential

pitfalls when using GT. It might be useful for potential users of GT to know of these and

what steps can be taken to avoid them.

The main advantage of GT is the rigour it can bring to qualitative research. However, it is not

always easy to maintain this rigor. The first area of problem concerns data collection. It is of

course essential to use theoretical sampling (see above) but a careful selection of initial

interviewees is necessary. Setting down the broad research questions the researcher seeks to

pursue is a good starting point. The researcher needs to be clear on the focus of their study –

does it comprise only accounting staff or also those with responsibility for financial decisions
and/or budgets, what levels within the organisation does it involve, does it extend beyond the

organisation to other stakeholders. An initial set of interviews with senior staff with

knowledge of the substantive area in the organisations will always pay dividends and help to

identify the first set of interviewees and the initial set of issues/questions to be pursued. Once

these initial interviews have been undertaken theoretical sampling can be pursued. Although

the extent of the interviews will be determined by theoretical saturation, this is unlikely to

occur with relatively few interviews. The minimum number of interviews in the projects I

have been involved in was about 50 and the maximum 150. One tip I found particularly

useful was to spend a few minutes after each interview in summarising the issues, themes etc.

that emerged. This is often the source of the initial coding and also helps with theoretical

sampling, identifying issues to be pursued, comparing and contrasting with prior interviews

and structuring future interviews. Gioia and Thomas (1996) similarly suggest to a 24-hour

rule for this initial analysis to be undertaken.

Data analysis is at the core of GT and there is a great deal of good advice in the various GT

texts to help researchers in carrying this out. It is however, never quite so straightforward in

practice. The most common problem is lack of confidence in the emerging coding and the

sense of losing one’s way. My advice would be to be patient and to discuss the coding with

others as much as feasible. I would also encourage researchers to not be too rigid or defensive

of their coding – be prepared to change until one is comfortable with the coding. Remember

GT is an iterative process so change is inherent to the process. Personally I found the use of

diagrams extremely helpful and would produce many different versions before coming to my

final interpretation. These were especially useful during axial and selective coding when the

relationship between categories and identification of the core category is undertaken. Finally,

write the ‘story’ of your findings in a few paragraphs. This is essentially the substantive GT
but it is easier for readers of your research (and probably yoursself!) to grasp than the detailed

coding and categorisation.

Other useful tips include the need to validate your findings by rehearsing them with a few of

the key interviewees. This is best done by telling the story referred to above and seeing if

they accept this as a fair interpretation. This provides some comfort with regards to validity

and reliability.

Don’t worry about strict adherence to methods/process, rather view GT as a guide to analysis

rather than be a slave to its methods. As can be seen above there is no agreement that one

particular approach to GT is better than another so there is no right or wrong way to

undertake the analysis as long as you can defend it.

Work with others whenever possible. Researchers will benefit from discussing their overall

approach, data collection and analysis with co-researchers and also with others not directly

involved with the research. Another major advantage of working with others is that is

provides some triangulation of interpretation and again provides some comfort with regards

to validity and reliability. GT is not an easy method and it is onerous and sometimes lonely so

sharing the burden wherever possible is advisable.

In all cases I have found the GT to be productive and informative. It is especially useful for

new researchers and doctoral students faced with the task of understanding complex social

settings of accounting and the task of collecting and analysing copious amounts of qualitative

data. I have also found it to provide a rigour often missing or at least not revealed in many

qualitative accounting publications. Like Suddaby (2006) I am aware that GT is certainly not

perfect and that there are many issues to be resolved by researchers using it but I have yet to

find a more convincing way of developing theories of accounting in complex social settings.

Issues and debates

Engaging with prior theory

In Strauss and Corbin’s approach, the researcher is encouraged to focus and clearly define the

topic and area of research by conducting an early literature review (Strauss and Corbin,

1998). Fendt and Sachs (2008) claimed that the researcher’s theoretical knowledge should be

seen as an advantage that can be utilised. Accordingly, many management researchers (e.g.:

Harris and Sutton, 1986; Eisenhardt and Bourgeois, 1988; Faltholm and Nilsson, 2010) have

adapted GT by starting their work with some prior specification of existing theory. Locke

(2001) claimed that the purpose of prior knowledge is “to bring more ordering and structuring

mechanisms into the analytic process” and to guard “against the real possibility of being

overwhelmed by the sheer volume of unstructured data” (p. 102). Besides, the idea of

conducting a manageable piece of research “without a clear research question and absent

theory simply defies logic” (Suddaby, 2006, p. 634) and is likely to produce an unsystematic

“mass of descriptive material waiting for a theory, or a fire” (Coase, 1988, p. 230). Suddaby

(2006, pps634-5) stresses that, ‘grounded theory is not an excuse to ignore the literature’ and

that ‘grounded theory research is always one of trying to achieve a practical middle ground

between a theory-laden view of the world and an unfettered empiricism. A simple way to

seize this middle ground is to pay attention to extant theory but constantly remind yourself

that you are only human and that what you observe is a of both who you are and what you

hope to see’. Strauss and Corbin (1990) indicated that a decision could even be made at the

early stage of the research, wherein the GT can be used to extend the existing theory by

applying it to a new context.

Nevertheless, a real concern regarding the issue of prior knowledge is that it might “force the

researcher into testing hypotheses, either overtly or unconsciously”, instead of making

discoveries about the phenomenon (Suddaby, 2006, p. 635). Besides, if relevant concepts of

the main problems are known prior to the research, there is no point in embarking upon

qualitative research. So, to tackle this issue, Corbin and Strauss (2008) advised the grounded

theorist to take a rather general approach in reviewing the literature.

However, taking a general approach is not the same as being driven by prior theory. Parker

(2014) discusses the critiques of Hambrick (2007) and Merino (1998) of always demanding

‘big’ macro theories to all research and the exhortations of Lewellyn (2003) and Humphrey

and Scapens (1996) to adopt alternative ways of theorising including creating and building

concepts induced from the field. He (Parker 2014 p23) summarises the case as follows, ‘in

contrast to this scientific credibility motivated macro-theory fixation, as argued in Parker

(2008), the qualitative tradition allows the researcher to both build new theory and modify

existing theory, through the inductive derivation of theory or through theoretical comparison

and critique (Humphrey and Scapens, 1996; Llewellyn, 2003; Parker and Roffey, 1997).’ GT

provides an excellent method for achieving this and Locke (2001) identifies two approaches

to theoretical development using GT, both of which have been used in accounting research.

Bamberger and Philips (1991), Gibbins et al. (1990), Cottingham and Hussey (1996), Parker

(2001 and 2002), Abdul Rahman and Goddard (1998) and Slagmulder (1997) have all

primarily relied on the empirical data alone to generate a theory. As outlined above, other

studies have made more extensive use of other ‘schools of thought’ to provide perspectives

on their research such as Coopey, Keegan and Emler’s (1998) use of structuration theory and

Covaleski et al’s (1998) use of a Foucauldian framework.

Is GT a methodology or just a method and does it matter anyway?

There are of course many different views on what constitutes methodology. In the

organisational and accounting literature perhaps the most influential researchers have been

Burrell and Morgan (1979) and Laughlin (1995). Burrell and Morgan clearly identify

methodology as a set of methods. However it is closely intertwined with the ontological and

epistemological stance of the researcher and their beliefs concerning human nature. This

leads to a basic dichotomy between subjective and objective assumptions underlying any

piece of research. Laughlin (1995) develops this further and incorporates a theoretical

element into his understanding of methodology, alongside a close relationship with the extent

to which prior theory is used and attitudes to societal change. Elharidy, Nicholson and

Scapens (2008 p.147) summarise his approach, ‘methodology concerns the “set of

spectacles” that determine the type of methods used for investigating the world (Laughlin,

1995); whereas methods are the specific techniques used to collect and/or analyse data.’.

In the GT literature there are also differences in views concerning the nature of methodology.

In their early work Strauss and Corbin (1990) describe GT as a qualitative research method

whereas Glaser (1992, p. 16) defines GT as ‘a general methodology’. Charmaz (2006)

contrasts objectivist with constructivist theories and identies Strauss and Corbin’s GT with

the former. However in their later work (Corbin and Strauss, 2015) the underlying

ontological assumptions of their approach to GT are recognised and clearly identified with

symbolic interactionism. They also accept the constructivist nature of theory development.

In the accounting literature Joannides and Berland (2008) clearly claim that GT is a

methodology rather than a method and that this is consistent with Strauss and Corbin (1998),

Parker and Roffey (1997) and Locke (2001). However, Elharidy, Nicholson and Scapens

(2008 p.147) agree with Strauss and Corbin’s definition of GT as a research method and

suggest that ‘treating GT as a methodology implies that it is a general philosophy about doing

research, coupled with a set of methods which are fundamentally influenced by its
ontological and epistemological assumptions’ and that ‘ the problem of confusing GT as a

methodology and GT as a method is that it can limit attention to the procedures (i.e. method),

rather than exploring the philosophical basis of the research (i.e. methodology).’ as a result,

there is a danger that the focus of the researcher could be on how to verify the emerging

codes, rather than on how to understand the nature of the phenomenon being studied.’ They

also note that if the researcher considers GT to be a research method, it can be used in

different methodologies including functionalist, interpretive, radical or even positivistic to

use Burrell and Morgan’s framework.

Where does all this leave the potential user of GT? Whether one considers GT to be a

methodology or method depends to a large extent on how one sees views methodology.

However in both cases, as in all research, the researcher needs to be aware of the issues and

of their own ontological and epistemological standpoint. They also need to ensure

consistency with the nature of the theory they are searching for. However Suddaby (2006,

p639) notes a word of caution that being aware of one’s epistemological position does not

justify dogmatism about conducting GT research and that ‘researchers should try to avoid

fundamentalist tendencies in how they approach and, more importantly, evaluate grounded

theory research’.

A final issue concerned with methodology is whether GT produces a theory at all. Again, this

depends on how one views theory. Charmaz (2005) suggests different ontological and

epistemological views lead to different views on theory. She contrasts positivistic theories

which seek causes and favour deterministic explanations with interpretive theories which call

for the imaginative understanding of the studied phenomenon. She further contrasts (p 131)

objectivist theory which ‘resides in the positivistic tradition and attends to data as real in and

of itself and does not attend to the processes of their production’ with constructivist theory
which she favours and ‘places priority on the phenomena of study and sees both data and

analysis as created from shared experiences and relationships with participants’.

Theories are often compartmentalised into either being deductively or inductively derived.

Suddaby (2006) suggests a more pragmatic approach be taken and that fundamentalists often

incorrectly describe quantitative approaches as necessarily deductive and GT as inherently

inductive. He appeals for recognition of Peirce’s (1903) approach of abduction which ‘is the

process of forming an explanatory’ and as the fallible “flash of insight” that generates new

conceptual views of the empirical world. Indeed, Strauss and Corbin noted that induction had

been overemphasized in GT research. They observed that whenever researchers

conceptualize data, they are engaging in deduction and that effective GT requires ‘an

interplay between induction and deduction (as in all science)’ (1998: 137). Again, the nature

of the theory developed by a GT researcher will depend largely on their own ontological and

epistemological standpoint.

Issues of validity and reliability

One of the principal criticisms of GT and of course to qualitative approaches in general is that

of ensuring validity and reliability. The functionalist mode of inquiry has dominated the

terrain of social science inquiry for such a long time that concepts such as ‘validity’ and

‘reliability’ are often construed within very narrow boundaries. To address this matter Strauss

and Corbin (1998 p. 266), for example, suggest a re-definition of these two canons of good

science to fit the realities and complexities of social phenomena, a view supported by a

substantial literature (e.g. Lincoln and Guba, 1985; Hammersley and Atkinson, 1995;

Hammersley, 1992; Miles and Huberman, 1994; McKinnon, 1988). From an interpretive

perspective validity reflects the extent to which a researcher’s account accurately or faithfully
represents the social phenomena that it seeks to describe, explain or theorise (Hammersley,

1992, p. 69). Reliability on the other hand reflects whether results of a study are consistent

with data collected, a matter that necessarily depends on a researcher’s analytical ability. The

literature on qualitative research offers suggestions on how to overcome, guard against or

minimise the impact of such threats on research output (e.g. Hammersley and Atkinson,

1995; Hammersley, 1992; Miles and Huberman, 1994; Strauss and Corbin, 1998; McKinnon,

1988). These comprise adopting sound strategies in research design, data collection, analysis

and presentation of research output.

Two strategies related to research design may be employed; a multi-site research design and a

discontinuous residence in the field. Multi-site/case research design allows not only greater

representativeness but also variation across cases. This strategy increases the range of

emergent concepts by exposing them to negative incidents across the sites. Discontinuous

residence in the field allows the researcher to ‘stand back and review critically’ (Miles and

Huberman, 1994, p. 264), a stance that permits construction of interim findings that are

subsequently host-validated. It also prevents over rapport that develops from continued long

residence in the field.

Ensuring accurate and faithful data capture is critical to credibility of any study – but more so

for a qualitative one. Audio recording of all pre-arranged interviews and writing of

observation notes enhances data credibility. Hard or soft copies of documentary data from all

sources should also be collected and copied.

Throughout open coding the analytical process of GT can be documented using logic

diagrams accompanied by the researcher’s theoretical notes – detailing the interpretation of

data as well as methodological pointers on what steps to take thereafter (Strauss and Corbin,

1990). Use of multiple sources [triangulation] of data should be built into the research design

as a strategy for data collection wherever possible. Multiple coding of data, incident variation
and host validation are three strategies that enhance validity and reliability of data analysis.

GT incorporates multiple coding of data through its three reflexive and iterative coding

procedures; open, axial and selective coding. Coding is initiated by a general reading or

listening followed by two or more open coding attempts. This way the researcher first

absorbs the data and obtains the first and then a second refined impression of the emerging

concepts. Incident variation seeks to examine concepts under a series of different conditions

and develop them across dimensional ranges. As suggested above host validation should also

be used. Miles and Huberman (1994) make a case for host validation using advanced interim

results as the stronger form of validation because the researcher knows more and better about

the research phenomenon. Schatzman and Strauss (1973) and Strauss and Corbin (1998) also

suggest testing and checking major propositions of the emergent theory against

understandings and experiences of hosts as a validity-enhancing course of action.

The general strategy in writing up and presentation of the research should be one of bringing

readers ‘back stage’ by including in output sufficient data through casing of lives and events

in organisations as well as extensive quotations from interview transcripts, documents and

observation notes. The objective is to create a text through which readers can follow the

logic, envision the process undertaken and weigh up the evidence made available to them.

The researcher should successfully show that his or her account is valid and reliable

Which approach to use?

Another issue which new GT researchers face is which approach to use. Should it be the

traditional approach advocated by Strauss and Corbin, the alternative developed by Glaser,

the constructivist approach suggested by Charmaz or the post modern, situationist, approach

of Clarke? The Sage Handbook on GT (2010) contains yet further developments and uses of
GT such as GT and action research, integrating GT and feminist methods, accommodating

critical theory, GT and the politics of interpretation, GT and diversity and GT and

ethnography. Unfortunately there is no real alternative in making a choice than obtaining an

understanding of all and selecting one that is most conducive to the researchers preferences

and underlying ontological and epistemological stance.

Once a general approach has been adopted the extent to which adherence to the detailed

methods and processes is made depends on personal preference. However, the researcher

must ensure adherence to the seven general principles of GT set out by Charmaz (2006) and

discussed above.

Getting published

There is no doubt that one of the reasons we see so few GT studies in accounting journals is

the difficulties in getting such work published. Aside from the resistance of a significant

number of editors and reviewers to inductive, qualitative and interpretive research in general

there is an additional resistance to GT. I suspect this is due to lack of familiarity and

understanding of GT which has led to some misunderstanding of its value an exaggerated

concern with some of the issues raised above. This is surprising given its complete

acceptance in other managerial and organisational disciplines. The most common problem I

have personally encountered is the relationship with prior theory and the imbued, essentially

functionalist research model held by many reviewers. The sorts of responses I have received

include an insistence on restructuring papers to include discussions of prior theory before the

GT is explained, suggestions that all references to GT be removed and replaced with just a

description of methods and even one response that insisted research cannot be undertaken

unless fully informed by prior theory. Most disappointingly this last response was from a
qualitative researcher. It should be said that I have also been advised by a reviewer

sympathetic to GT to remove all references to prior theory entirely. It should be noted that all

the papers were eventually published in good journals but compromises had to be made. It

can only be hoped that such responses become less frequent as reviewers become more

familiar with GT and accept its legitimacy and that journal editors become more supportive.

However, the problem of getting published is far from being only due to reviewers.

Researchers themselves must ensure the rigour of their research and the readability of their

papers. The review of management GT papers undertaken by Suddaby (2006) is of particular

use to GT researchers. Although supportive of GT, Suddaby provides an excellent summary

of the pitfalls to be avoided. Most of the GT texts provide more general guidance to writing

papers and getting published (e.g. Corbin and Strauss 2015, chap 17, Charmaz 2006 chap 7).

My personal advice would be to ensure clarity over the methods used, to provide clear

explanation of the audit trail from raw data to final theory, to concentrate on the core

category, to write the story of the emergent theory, to engage in some way with prior theory

to move the substantive theory towards a more formal theory and to clarify the theoretical

and empirical findings.

Future of GT in accounting research

Given the acceptance and preponderance of GT research in other social science disciplines it

seems inconceivable that GT will not become more extensively used in accounting research.

GT addresses many of the calls for developing interpretive, management accounting theories

(Elharidy, Nicholson and Scapens, 2008), theories from case studies (Humphrey and Scapens

, 1996), accounting theory development using alternative approaches (Llewellyn, 2003) and

more qualitative accounting research (Ahrens and Chapman 2006).

Glaser (2010) suggests the spread of GT has been partly due to its usefulness in an

increasingly globalised world characterised by culturally diverse environments where

differences cannot be imagined or conjectured but must be discovered. He further suggests

that ‘as a consequence of cultural diversity, more and more researchers and users of the more

evidentiary, preconceived formulated research have become disaffected with their data

collection, their findings, what they should find, and whatever hypotheses should be tested.

Smoldering disaffection has grown as findings are seen to be beside the point, irrelevant,

moot, and unworkable’ (Glaser, 2010, p 5-6). As outlined above GT researchers are already

exploring its use alongside action research, integrating GT and feminist methods,

accommodating critical theory, the use of GT and the politics of interpretation, GT and

diversity and GT and ethnography. All these developments are likely to be relevant and of

interest to accounting researchers.

Parker and Roffey’s (1997, pps 243-4) summation of the potential advantages of GT to

accounting research is as relevant today as it was nearly 20 years ago, `Grounded theory, as a

potentially valuable part of the qualitative interpretive field research tradition, offers the

prospect of contributing important dimensions of knowledge to accounting research….It

offers the prospect of providing useful confirmation or disconfirmation of the applicability of

pre-existing theories as well as the possibility of offering new theoretical developments.

These developments may carry the advantage of environmental sensitivity, given grounded

theory’s emphasis on grappling with the multiple complexities of observed environments. It

offers a systematic framework for the study of accounting systems and management practices

in their social and organizational contexts. …It also offers the possibility of providing more

general observations that may have ramifications beyond the particular case being studied

(Strauss, 1987).’
The traditional areas of GT accounting research are those associated with social settings such

as the everyday accountant and manager (Parker and Roffey 1997).Within these settings GT

has been exhorted to undertake interpretive management accounting research by Elharidy,

Nicholson and Scapens ( 2008) and Joannides and Berland (2008). There are many specific

areas where GT is particularly appropriate. These include areas where social processes are

most active such as the development and implementation of new accounting processes,

accounting and organizational/cultural change and accounting and ethics (Parker and Roffey

1997). With the failure of existing economic models to understand the recent (and possibly

the impending) financial crisis, GT offers a method for a deeper understanding of financial

market decision makers. Behavioral finance has already made the move away from abstract

theories of rational decision making and GT studies have the capacity to expand this

understanding.

GT is particularly useful in areas with little prior research. With academic accounting

research still at its early stages compared to other disciplines, such areas abound. Examples

include accounting regulation, accounting and governance, the ‘everyday auditor’,

development and implementation of new accounting techniques and processes, accounting

and IT, accounting in developing and emerging countries, local government accounting,

accounting and politicians and many more.

However, GT research does face many serious challenges. In many countries, perhaps the

biggest threat to GT is the emphasis on producing publications quickly and in quantity.

Compared to quantitative research it is resource and time consuming with a relatively limited

output to input ratio. Allied to the hesitancy of accounting journals to accept GT the risk for

new accounting researchers to adopt GT is indeed significant. It can only be hoped that the

academic community will accept the huge potential GT has for accounting research and

encourage its adoption.

Conclusion

For several decades accounting researchers have called for understanding the subject in its

social, societal and political setting (Tomkins and Groves 1983, Hopwood 1983, Humphrey

and Scapens 1996, Parker and Roffey 1997, Llewelyn 2003, Ahrens et al 2007). Most

academics would agree that theory is essential for developing an understanding of

phenomena. Again over the last three decades a number of calls for theories of management

accounting have been made (Humphrey and Scapens 1996, Llewelyn 2003 Malmi, and

Granlund, 2009). These calls have common a common theme, that where theory has been

utilised by accounting researchers it has been imported from elsewhere (economic, social

organisational theories) or is merely normative or descriptive. As long ago as 1996

Humphrey and Scapens (1996, p100) noted that ‘social theories have brought the “political”

and the “social” into the realms of accounting knowledge, but it has not produced significant

insights into the intricacies, diversities or contradictions of accounting practice in

contemporary organizations. As explanations have largely resided in social theories

themselves, rather than in (theoretically informed) observation, accounting research has

struggled to explain the differential nature of day-to-day accounting practices.’

Malmi, and Granlund (2009) noted that there are few cases of theories of management

accounting, rather they are theories about management accounting. They go as far as to claim

that there is no theory unique to management accounting that is currently considered

scientific by the international research community. Humphrey and Scapens (1996 p 100)

called for the use of qualitative case study research to develop theories and that, ‘the potential

of case studies will be enhanced by bringing conversations about theory more explicitly into

the accounting research process; in particular, by recognizing the dynamics of the interaction
between theory and observation’. However they were less clear on how this might be

achieved in practice. Hopefully this chapter has convinced potential users that GT can

address these lacunae and deliver better and deeper theories of accounting.

References

Abdul-Rahman, A.R. and Goddard, A. (1998), “An interpretive inquiry of accounting

practices in religious organisations”, Financial Accountability & Management, Vol. 14 No. 3,
pp. 183-200.

Ahrens, T. (1996), “Styles of accountability”, Accounting Organizations and Society, Vol.21,

No.2/3 pp.139-173.

Ahrens, T. and Chapman, C. (2006), ‘Doing qualitative field research in management

accounting: positioning data to contribute to theory’, Accounting, Organizations and Society,
31(8),pp. 819– 841).

Allen, L., (2010), ‘A Critique of Four Grounded Theory Texts’, The Qualitative Report
Volume 15 Number 6 November 2010 1606-1620

Bamberger, P and Philips, B, (1991), “Organisational Environment and Business Strategy:

Paralell Versus Conflicting Influences on Human Resource Strategy in the Pharmaceutical
Industry”, Human Resource Management, Summer.

Beattie, V.A., Fearnley, S. and Brandt, R., (2004), ‘A Grounded Theory Model of Auditor-
Client Negotiations’, International Journal of Auditing, 8, 1-19.

Burrell, G and Morgan, G (1979), Sociological Paradigms and Organisational Analysis:

Elements of the Sociology of Corporate Life, Ashgate Publishing Ltd.

Charmaz, K. (2006), Constructing Grounded Theory, A Practical Guide through Qualitative

Analysis, London: SAGE Publication Ltd.

Charmaz, K. (2014) Constructing Grounded Theory, A Practical Guide through Qualitative

Analysis, Second ed., London: SAGE Publication Ltd.

Clarke, A., (2003), ‘Situational Analyses: Grounded Theory Mapping After the Postmodern
Turn’, Symbolic Interaction, Volume 26, Number 4, pages 553–576.

Clarke, A. (2005). Situational analysis: Grounded theory after the postmodern turn.
Thousand Oaks, CA: Sage.

Coase, R. (1988), The firm, the market, and the law. Chicago: University of Chicago Press.
Coopey,J., Keegan, O., and Embler, N., (1998), ‘Managers’ Innovations and the
Structuration of Organisations’, Journal of Management Studies, 35, 264-84.

Corbin, J. M. (1998). Comment: Alternative interpretations-valid or not? Theory and

Psychology, 8(1), 121-128.

Corbin, J. and Strauss, A. (1990), “Grounded theory research: procedures, canons, and
evaluative criteria”, Qualitative Sociology, Vol. 13 No. 1, pp. 3-21.

Corbin and Strauss (2008) Basics of Qualitative Research: Grounded Theory Procedures
and Techniques, 3rd ed. Sage, London.

Corbin and Strauss (2015) Basics of Qualitative Research: Grounded Theory Procedures
and Techniques, 4th ed., Sage, London.

Cottingham, J and Hussey, R., (1996), “A Grounded Theory Study of Related Party
Tansactions”, Proceedings of the ICAEW Beneath The Numbers Conference, January,
Portsmouth.

Covaleski, M. and Dirsmith, M. (1983), “Budgeting as a means of control and loose

coupling”, Accounting, Organizations and Society, Vol. 8 No. 4, pp. 323-40.

Covaleski, M. and Dirsmith, M. (1984), “Building tents for nursing services through
budgetarynegotiation skills”, Nursing Administration Quarterly, pp. 1-11.

Covaleski, M., Dirsmith, M., Heian, J., & Samuel, S. (1998). ‘The calculated and the avowed:
Techniques of discipline and struggles over identity in Big Six public accounting firms’,
Administrative Science Quarterly, 43 , 293–327.

Eisenhardt, K., & Bourgeois, L. J., (1988),‘Politics of strategic decision making in high-
velocity environments: Toward a midrange theory’, Academy of Management Journal, 31:
737- 770.

Efferin, S. and Hopper, T. (2007), “Management control, culture and ethnicity in a Chinese
Indonesian company”, Accounting, Organizations and Society, Vol. 32 No. 3, pp. 223-62.

Elharidy A, Nicholson B, and Scapens, R. (2008),"Using grounded theory in interpretive

management accounting research", Qualitative Research in Accounting & Management, Vol.
5 Iss 2 pp. 139 - 155

Faltholm, Y. and Nilsson, K. (2010), Business Process Re-engineering and Balanced

Scorecard in Swedish Public Sector Organizations: Solutions for problems or problems for
solutions? , International Journal of Public Administration, Vol. 33, pp. 302 – 310.

Farneti, F. (2009), Balanced scorecard implementation in an Italian local government

organization, Public Money & Management, September 2009, pp. 313 – 320.

Fendt, J. & Sachs, W. (2008), Grounded Theory Method in Management Research: User’s
Perspectives, Organizational Research Methods, Vol. 11, No. 3, pp. 430 – 455.
Gibbins, M., Richardson, A. and Waterhouse, J. (1990), “The management of corporate
financial disclosure: opportunism, ritualism, policies, and processes”, Journal of Accounting
Research, Vol. 28 No. 1, pp. 121-43.

Gioia, D and Thomas, J (1996) Identity, Image, and Interpretation: Sensemaking during
Strategic Change in Academia, Administrative Science Quarterly, Vol. 41, No. 3, pp. 370-
403.

Glaser, B. (1992), Emergent vs. Forcing: Basics of Grounded Theory Analysis, Sociology
Press, Mill Valley, CA.

Glaser, B. (2010), ‘The Future of Grounded Theory’, The Grounded Theory Review (2010),
vol.9, no.2 pp.1-14.

Goddard, A. (2004), “Budgetary practices and accountability habitus: a grounded theory”,

Accounting, Auditing & Accountability Journal, Vol. 17 No. 4, pp. 543-77.

Goddard, A. (2005), “Accounting and NPM in UK local government contributions towards

governance and accountability”, Financial Accountability & Management, Vol. 21 No. 2,
pp. 191-218.

Goddard, A. and Assad, M.J. (2006), “Accounting and navigating legitimacy in Tanzanian
NGOs”, Accounting, Auditing & Accountability Journal, Vol. 19 No. 3, pp. 377-404.

Goddard A.R. and Tillmann K, (2008), ‘Strategic management accounting and sense-making
in a multinational company’, Management Accounting Research, Volume 19, Issue 1, March,
Pages 80-102.

Goddard A R and Assad,M, (2010), ‘Stakeholder salience and accounting practices in

Tanzanian NGOS’, International Journal of Public Sector Management, Vol 23.No 3. 2010.

Goddard A R and Issa, S. (2015) ‘Accounting Practices in Tanzanian Local Government

Authorities: Towards a Grounded Theory of Manipulating Legitimacy’, Research in
Accounting in Emerging Economies, forthcoming.

Goddard A R and Malagila J. (2015) ‘Public Sector External Auditing in Tanzania: A Theory
of Managing Colonising Tendencies’, Research in Accounting in Emerging Economies,
forthcoming.

Goddard A, and Mkasiwa T (2015), ‘Accounting Changes and Budgeting Practices in

Tanzanian Central Government: ‘Struggling for Conformance’, Journal of Accounting in
Emerging Economies, forthcoming.

Goddard A, Assad, M., Issa S., Malagila J. and Mkasiwa T., (2015) ‘ The ‘Two Publics’ and
Institutional theory– A Study of Public Sector Accounting in Tanzania’, Critical Perspectives
on Accounting Journal, forthcoming.

Glaser, B. and Strauss, A. (1967), The Discovery of Grounded Theory: Strategies for
Qualitative Research, Aldine Publishing Co., Chicago, IL.
Gurd, B. (2008), “Remaining consistent with method? An analysis of grounded theory
research in accounting”, Qualitative Research in Accounting & Management, Vol. 5 No. 2,
pp. 122-138.

Hambrick, D.C. (2007), “The field of management’s devotion to theory: too much of a good
thing?”, The Academy of Management Journal, Vol. 50 No. 6, pp. 1346-1352.

Guba, E and Lincoln, Y (1998) Competing Paradigms in Qualitative Research in Denzin, N

and Lincoln, Y. (Eds.) ,The Landscape of Qualitative Research: Theories and Issues, SAGE
Publications, 1998, pp. 195-220.

Hammersley, M (1992) What’s Wrong with Ethnography? Methodological explorations,

Routledge.

Hammersley, M. and Atkinson, P. (1995), Ethnography: Principles in Practice, Routledge,

London.

Harris, S. G. & Sutton, R. I. (1986), Functions of parting ceremonies in dying organizations,

Academy of Management Journal, Vol. 29, pp. 5 – 30.

Humphrey, C. and Scapens, R. (1996), ‘Theories and case studies of organizational

accounting practices: limitation or liberation?’, Accounting, Auditing and Accountability
Journal , 9(4), pp. 86–106.

Kendall, J. (1999), Axial Coding and the Grounded Theory Controversy, Western Journal of
Nursing Research, Vol. 21, No. 6. pp. 743 – 757.

Laughlin, R. (1995), “Empirical research in accounting: alternative approaches and a case for
middle range thinking”, Accounting, Auditing & Accountability Journal, Vol. 8 No. 1,
pp. 63-87.

Lightbody, M. (2000), “Storing and shielding: financial behaviour in a church organization”,

Accounting, Auditing & Accountability Journal, Vol. 13 No. 2, pp. 156-74.

Lincoln, Y and Guba, E (1985) Naturalistic Inquiry, Sage Publications.

Locke, K. (2001), Grounded Theory in Management Research, Sage, London.

Llewelyn, S. (2003) What counts as ‘theory’ in qualitative management and accounting

research, Accounting, Auditing and Accountability Journal, 16(4), pp. 662–708

Malmi, T., & Granlund, M. (2009), ‘In Search of Management Accounting Theory’,
European Accounting Review,18(3), 597–620.

McKinnon, J (1988) Reliability and Validity in Field Research: Some Strategies and Tactics,
Accounting, Auditing & Accountability Journal, Vol. 1, No. 1, pp. 34-54.

Merino, B.D. (1998), “Critical theory and accounting history: challenges and opportunities”,
Critical Perspectives on Accounting, Vol. 9, pp. 603-616.
Miles, M and Huberman, A (1994) Qualitative Data Analysis, 2nd Edition, SAGE
Publications.

Parker, L.D. (2012), “Qualitative management accounting research: assessing deliverables

and relevance”, Critical Perspectives on Accounting, Vol. 23, pp. 54-70.

Parker, L.D. (2001), “Reactive planning in a Christian bureaucracy”, Management

Accounting Research, Vol. 12, pp. 321-56.

Parker, L.D. (2002), “Budget incrementalism in a Christian bureaucracy”, Management

Accounting Research, Vol. 13 No. 1, pp. 71-100.

Parker, L.D. (2008), “Interpreting interpretive accounting research”, Critical Perspectives on

Accounting, Vol. 19 No. 6, pp. 909-914.

Parker L. , (2014),"Qualitative perspectives: through a methodological lens", Qualitative

Research in Accounting & Management, Vol. 11 Iss 1 pp. 13 – 28.

Parker, L.D. and Roffey, B.H. (1997), “Back to the drawing board: revisiting grounded
theory and the everyday accountant’s reality”, Accounting, Auditing and Accountability, Vol.
10 No. 1, pp. 212-47.

Peirce, C. S. 1903. The essential Pierce: Selected philosophical writings, vol. 2.

Bloomington: Indiana University Press.

Robrecht, L.C. (1995). Grounded Theory: Evolving Methods. Qualitative Health Research
,Vol. 5 p169.
Schatzman, L. (1991), “Dimensional analysis: notes on an alternative approach to the
grounding of theory in qualitative research”, in Maines, D.R. (Ed.), Social Organization and
Social Process, Aldine de Gruyter, New York, NY.

Schatzman, L., & Strauss, A. L. (1973). Field research. Englewood Cliffs, N.J.: Prentice-
Hall, Inc.

Slagmulder, R. (1997), “Using management control systems to achieve alignment between

strategic investment decisions and strategy”, Management Accounting Research, Vol. 8
No. 1, pp. 103-39.

Solomon, J.F and Solomon, A, (2006), “Private social, ethical, and environmental disclosure”
Accounting, Auditing and Accountability Journal, Vol.19, No,4 pp.564-591.

Strauss, A.L. (1987), Qualitative Analysis for Social Scientists, Cambridge University Press,
Cambridge.

Strauss, A.L. and Corbin, J. (1990), Basics of Qualitative Research: Grounded Theory
Procedures and Techniques, Sage, London.

Strauss, A and Corbin, J (1998) Basics of Qualitative Research – Techniques and Procedures
for Developing Grounded Theory, 2nd Edition, SAGE Publications.
Suddaby, R. (2006), “From the editors: what grounded theory is not”, Academy of
Management Journal, Vol. 40 No. 4, pp. 633-42.

Wickramasinghe, D., Hopper, T. and Rathnasiri, C. (2004), “Japanese cost management

meets Sri Lankan politics: disappearance and re-appearance of bureaucratic management
controls in a privatised utility”, Accounting, Auditing & Accountability Journal, Vol. 17
No. 1, pp. 85-120.