Summary Applied Research Method

Lecture 1: Introduction

What is Applied Research? Organized, systematic, data-based, critical, objective, inquiry,

or investigation into specific problem with purpose of finding answers/solutions.

Why does it matter? Help managers make good decisions and solve specific problems
(Evidence-based-management)

 Where does the problem lie?

 What are the intervening factors?
 What types of information need to be gathered?
 How to make use of that information?
 How we implement the results of the process?

Why should you learn how to do it? Main advantages:

 Higher engagement in problem-solving

 Less time in understanding nature of the problem
 Implement/change recommendations
 Lower costs (compared to external teams)

Characteristics of scientific (good) research:

 Purposiveness
 Rigor
 Testability
 Replicability
 Precision and confidence
 Objectivity
 Generalizability
 Parsimony (choosing simplest explanation)

Two types of methods:

Hypothetico-deductive method Inductive method (Induction):

(Deduction):

 Starts with a general theory of a  Starts with the observation of

topic that is to be tested specific phenomena and attempts to
 Narrow toward specific generalize
hypotheses and tests those o specific to general

 Test of specific hypothesis  Why does this happen?

 Often quantitative  Provisional explanation - hypothesis
 Often qualitative

1
Important elements to consider when writing a proposal:

 Purpose - Form will vary according to situation

 Audience - Details to be adjusted to target audience (scientists, investors?)
 Characteristics - Clarity, conciseness, coherence. Focus on the ‘right’ aspects.

Ethical considerations:

 Why is this research project worth doing?

 How does the organization benefit from this project?
 What impact (positive + negative) could/does my research have on the organization?
 Do you have the skills and resources to carry out this research project?
 Have you informed all the employees of the research project? Why not?
 Do you explain the purpose of your research to the participants? Why not?
 Are participants given the opportunity to decline participation?
 Are participants able to withdraw their consent at any point? How?
 Does the research cause you to have access to sensitive information? How will you
ensure the confidentiality of this information?
 How will you ensure individual respondents cannot be identified from any research
reports or papers that are produced?
 Are there any possible negative effects (long or short term) on your participants
(including any physical or psychological harm)?
 How will you report back from the research to your participants?
 Where ethical dilemmas have arisen, what steps have you taken to resolve these?

2
Lecture 2: The Problem

5 core principles of entrepreneurship

1. Start with your means and take action (vs wait for the perfect opportunity)
2. Set affordable loss (vs focus on the attractiveness of the upside)
3. Leverage contingencies and embrace surprises (vs over commitment to existing goals)
4. Form partnerships with people really committed (vs me against the world or team up
with partially committed people)
5. Create opportunities with what is under your control (vs focus on predicting the future
or finding the optimal opportunity)

Problem - Any situation where a gap exists between an actual and a desired ideal state.

1. Identification of the broad management problem 3 steps in the research process

2. Preliminary research  Closely associated with Kurt Lewin’s
3. Definition of the research problem action research paradigm

Kurt Lewin’s action research paradigm:

The Kurt Lewin action research paradigm is a problem-solving and change management
approach that emphasizes collaboration between researchers and practitioners to identify and
address organizational or social issues. This approach was developed by Kurt Lewin, a social
psychologist who is widely considered as one of the pioneers of social psychology and action
research.

The Lewinian action research paradigm involves a cyclical process of planning, acting,
observing, and reflecting. In this process, researchers work closely with practitioners or
members of the organization to identify problems, generate hypotheses, develop
interventions, implement the interventions, and then evaluate their effectiveness. This
iterative process allows for continuous learning and improvement. In other words, this
iterative nature ensures that the process remains responsive to changing conditions and
emerging issues, ultimately leading to more effective and sustainable solutions.

3
 1. Identification of the problem

Broad problem must be transformed into a feasible topic for research

 Making it more specific and precise
 Setting clear boundaries

Problems vs symptoms of problems: symptoms are not the real problem  the 5 whys will
lead you to the real problem.

EXAMPLE: My best employees are leaving the organization.

 Why? Because they are not satisfied with their jobs.
 Why? Because they do not find a challenge in their jobs.
 Why? Because they do not have control over their work.
 Why? Because they do not have a lot of influence over planning, executing, and
evaluating the work they do.
 Why? Because we have been reluctant to delegate.
Always ask why this problem occurs to see if the root problem you identified is connect to the broad problem or not?

2. Preliminary research

Background information on context Information on topic/area Competition

 Organization (origin, history, size,  Key players, possible

purpose and values, location, resources,  What are the current new entrants?
institutional relationships, finances) understandings & viewpoints?  What is their
 Textbooks, journal, articles, competitive
 Environment conference proceedings, and advantage?
 Primary (first-hand) vs secondary other published and
 Allows you to build on
(reports, statistics) data unpublished materials
what’s known, increase
 CAREFUL with sources
 Timeliness (when were the data accuracy and clarity,

4
 collected)  Allows you to build on what is remain focused on what is
 Accuracy (who, how and why already known, increase accuracy important.
were they collected) & clarity, remain focused on what
 Relevance (Is it applicable) is important.
 Costs

3. Definition of problem statement

Exploratory Descriptive Causal

 Not much in known  Data characterizes objects, events, or  Does one cause the other?
 Existing research is unclear situations  Tries to find root of the
 Highly complex topic  Helps understand given segment problem and isolate cause
 Not enough theory available  Offers additional ideas (from other potential
 Useful for simple decisions explanations)
 Focuses on why and usually
qualitative research  Can be quantitative (what?) or  Quantitative (experiments)
qualitative (how?)

Lecture 3: Research Design and Process

Defining targets of your applied research:

Who knows? Who cares? Who can?

Who can provide information, Who is involved, part of the Who can do something about the solution?
shed light on problem system? Are there third parties involved?

Critical literature review: Be aware of: Do not forget:

 Helps you narrow your broad  Purposely  Plagiarism makes it difficult for the reader to
problem misrepresenting verify whether your claims about other
 Reduces waste of effort the work of authors and sources are accurate.
 Clarifies terminology and other authors  You need to make your position in this debate
defines key terms  Plagiarism clear by designating the authors whose work
you are building on or whose ideas you are
challenging.

5
 Research Design

Research Strategies:

Ethnography (Qualitative) Case studies (Qualitative)

 Roots in anthropology  About a specific object, event, or activity

 Close observation of daily life  The target can be the individual, group,
 Involves immersion in the group organization, situation, or event
 Investigation in real life context

Interviews (Grounded theory) (Qualitative) Action research (mixed)

 Inductively deriving theory from the data  Study of planned changes

 Develops theory as it emerges  Continuous process
 Interviews (and more) to collect data  Problem – solution – effects – new problem

Experiments (quantitative) Surveys (quantitative)

 Examines causal relationships  Asks multiple questions

 Manipulation on the independent variable  Describe, compare, or explain knowledge,
 Should include a control/comparison group attitudes, behaviors

6
 OTHER RESEARCH ELEMENTS:

 Researcher interference: Varies substantially

 Study setting: Natural vs artificial settings
 Unit of analysis: Individuals, dyads, groups, organizations, industries, countries
 Time horizon: Cross-sectional vs time-lagged / longitudinal

Contrived means artificial setting that was specifically set up for the purpose of study
Cross-sectional is the observation at certain point in time

The Importance of Triangulation:

Mixed Methods Approach Combination of qualitative and quantitative research methods

Method triangulation Multiple collection and analyzation methods

Data triangulation Several sources/time periods

Researcher triangulation Multiple researchers

Theory triangulation Multiple theories/perspectives

Design does not matter if population is not correctly targeted (i.e., if data are not collected from the people,
events, or objects that can provide the correct answers to solve the problem). If sample is not representative (if
sample values fall outside of population parameters, overrepresented or underrepresented)

Sampling

 Population - entire group or people, events or things of interest research wishes to

investigate
 Element - single member of population
Elements  Sample - subset of population
 Sampling unit - single object that can be selected with certain probability during the
sampling process, example: household, organization
 Subject - single member of sample; a participant of the actual study

1. Define the population

a. In terms of elements, geographical boundaries, time
b. Objective and scope of the study play a crucial role

2. Determine sample frame

a. Accurate representation of all elements in population from which sample is drawn (if
bad we get bigger coverage error)

3. Determine sampling design

Sampling a. How important is generalizability?

7
 b. Probability vs non-probability sampling

Probability Sampling:

Simple Random

o Unrestricted sampling (also known as simple random sampling) - every element with
equal chance of selection. Often use computers for generation of random numbers

Complex
Process
o Systematic sampling - every nth element

o Stratified random sampling (to be sure every nationality, section, gender, study is
represented) – process of stratification or segregation, followed by random selection of
subjects from each stratum. Population is first divided into mutually exclusive groups that
are relevant, appropriate, and meaningful.

o Cluster sampling – samples gathered in groups or chunks of elements that, ideally, are
natural aggregates of elements in the population.

o Doubled sampling - initial sample for preliminary analysis followed by a subsample to

examine in more detail.

Non-Probability Sampling:

o Convenience sampling – process of stratification or segregation followed by random

selection of subjects from each stratum. Population is first divided into mutually exclusive
groups that are relevant, appropriate, and meaningful.
o Purposive sampling – obtain information from specific target group. Specific types of
people who can provide info

4. Determine appropriate sample size

It depends on the research objective, the extent of precision desired (the confidence interval), the
acceptable risk in predicting the precision (confidence level), the amount of variability in the
population itself, the cost and time constraints (few costs for ex), the size of pop, the accessibility.

8
 5. Execute the process

 Implement the decisions & be careful with nonresponse error

Strategies to gain access to your sample

 Allowing yourself sufficient
 Using existing and developing new contacts
 Providing a clear account of purpose and type of access required
 Overcoming concerns (time, resources, sensitivity, confidentiality)
 Highlighting possible benefits to the individual/organization
 Using suitable language
 Facilitating replies
 Developing access incrementally
 Establishing credibility

Mistakes in  Ask questions to people giving you the answer you are looking for
 Research based on friends is not replicable
market
 Size or methodology is flawed
research  Bias your own research to give you the answer you want
 Ignore intangibles such as brand, customer service, channel

 Privacy
 Voluntary nature of participation and right to withdraw partially or completely
Ethical  Consent and possible deception
considera-  Maintenance of the confidentiality of data provided
 Reactions of participants of the way you seek to collect data (embarrassment, stress etc.)
tions  Effects on participants of the way you use, analyse and report data (embarrassment, stress)
 Behaviour and objectivity of you as a researcher

Lecture 4: Data Collection Methods (Interviews)

Research Paradigms – Comparison

Qualitative : Based on the existence

of many realities rather than a single
reality.
Behavior is complex and cannot be
predicted by predefined probabilities.
The behavior of society is not like a
scientific variable that can be
controlled. Here, methods of
interpretation or understanding
meanings are used to understand
behavior.

9
Quantitative : A unique reality that can be measured and understood. Therefore, quantitative approaches are
used to quantify this reality. Real survey, using a rigorous approach of systematic study of data sources.

Research Paradigms - Data Quality Criteria

Deductive research : The deductive

method is an explanation based on a
cause. In this case, the student starts
with one or more working hypotheses
to be validated or invalidated. It’s a
technique that starts with one or more
working hypotheses and works
towards explaining these hypotheses.

From general assumptions, this

method allows the student to
understand particular elements (from
the general to the specific).

Inductive research : The inductive method is an explanation based on a fact. It is a working method that starts
from facts, from real and observable raw data, and moves towards their explanation.

From the particular phenomena observed in the field, the researcher can understand a general phenomenon. We
go from the particular to the general.

Choosing Data Collection Methods

 Objectives/purpose – what are the questions we need to answer

 Skills and expertise – be aware of your limitations
 Cost/budget – sometimes most effective method is not possible
 Time – what is the amount of time available
 Availability – what types of data are readily available

Qualitative vs. Quantitative

Qualitative methods: Exploring and understanding people’s needs + Testing reactions to

concepts + Working out what the real issues/problems are

Quantitative methods Standardized: allows quantification (% individuals that …) + Large

numbers / robust measurement

10
 When to use them?
▪ What are the problems/opportunities? QUALITATIVE
▪ What are the options to solve a problem? QUALITATIVE
▪ What should the decision be? QUANTITATIVE

Qualitative Interviews

Approaches Pros and Cons

 Conversations and storytelling – listening

in to the conversation and general gossip
and story telling
 Semi-structured interviews – topics, issues
and participants determined in advance
 In-depth interviews – less structure, builds
on responses
 Group interviews – when group dynamic
leads to data
 Focus groups – role of the moderator is
critical, use technique to elicit data
(stimulus)

Planning and conducting Types of Questions

 Determine objectives and topics for  Open questions – participants define/describe a situation
discussion or event (e.g.: How has corporate strategy changed over
 Identifying and approaching interviewees the last 5 years?)
 Permission  Probing questions – used to explore responses that are
 Arranging interviews (time, topics) significant to the research topic (e.g.: what external
 Conducting interviews factors caused the corporate strategy to change?)
 Specific and closed questions – used to obtain specific
info or confirm fact or opinion (e.g.: How did David react
to the change in corporate strategy?)

Probing Questions - Tactics Difficulties

 Silence
 Repeating answer
 “So, what i hear you saying.”
 “I’m not quite sure I understood… Could
you…?”
 “Tell me more about”
 “Give me an example”
 “Go over that again”

11
 “Anything else?”

Concept Test

 “What do you find attractive about the product/service described?”

 “What do you dislike?”
 “What would you make better?”
 “Do you think it is feasible?”
 “Do you have any suggestions for increasing the chance of success?”

Key Take Aways Interviews

 Validity, reliability, and generalizability are important

 Think before you choose the data collection methods
 Interviews: Understand the multiple approaches | Pay attention to how you question
and probe | Find the key themes | Anticipate and prepare for barriers

12
 Lecture 5: Data Collection Methods (Questionnaires)

Questionnaire goal - collect information from large numbers of people in standardized form

How to collect:

Questionnaire types Pros and Cons

 Personally
o Advantages: establish rapport, clarification is
possible, less expensive in groups & higher
response rate.
o Disadvantages: might introduce bias, time and
effort.

 Mail
o Advantages: anonymity, wide geography can be
reached, convenience for respondents, token gifts
can be included.
o Disadvantages: low response rate, no
clarification possible.

 Online
o Advantages: easy to administer, global reach,
very inexpensive, fast delivery, convenience for
respondents, automatic processing.
o Disadvantages: requires computer literacy,
sampling issues, high non-response rate.

Things to keep in mind Types of scales

 Content and purpose – subjective feelings or  Dichotomous – yes/no

objective facts (scales vs single questions)
 Category or nominal – forms of transport: cars,
 Language and wording – be aware of the level of trains…
understanding of respondents (cultural
interpretation and frames of reference)  Semantic differential – introverted ———
extroverted
 Sequencing of questions – funnel approach (from
general to specific questions), check for ordering  Numerical – introverted - 54321 – extroverted
effects
 Likert – 1 strongly agree …, 5 strongly disagree
 Personal information – beginning or at the end? If
sensitive, use intervals  Constant sum – divide 100 points by the
following items

13
 Type and form of questions Make Sure
o Open vs closed – choices, possible  General appearance (attractive and neat, provide
alternatives
o Positively vs negatively worded items – mix a good introduction)
them/ careful with double negatives  Organize and align questions
o Recall-dependent questions – need to
reduce/account for bias  Provide instruction + guidance
o Length of questions – shorter are preferable  Review + test the survey
 Take advantage of existing systems (Qualtrics)
 Avoid:

o Double-barrelled questions – involves two

subparts
o Ambiguous questions
o Leading questions
o Social desirability

Translating Measures

Key take-aways: Questionnaires:

 Different collection methods (pros/cons)
 Validity, reliability, and generalizability
 How to frame statements/ask questions
are important
 Think before you choose the data  Multiple types of scales available
collection methods  Looks matter! Appearance issues
 Be careful with translation methods

14
 Lecture 6: Data Collection Methods and Action Research

Other Data Collection Methods

 Only if relevant for the project, examined before data collection (cost reduction).

 Useful for:
o Identifying problem/setting objectives – shapes the primary data collection
o Developing an approach to the problem – to define and refine the approach
Secondary o Formulating the appropriate research design – identify and isolate key variables
data o Answering certain research questions – often resolves certain research questions
o Helping interpret primary data – provide additional or different insights into data
collected via primary methods

 Criteria:
o Methodology – check the reliability and validity of the data
o Error/accuracy – use several sources / triangulation
o Date of data collection – prediction is very difficult especially if it is about the future
o Purpose/content of data collection – relevance, appropriateness to your problem
o Dependability/source credibility – Hidden agendas, trustworthiness, reputation

 Advantages: relies on what people do, not what they feel or say, helps identify subtle and
hidden issues, richer and more detailed, good preliminary data.

 Techniques of data collection: cameras, microphones, activity meters, written records

(diaries, clipboards), written case notes.

 Characteristics of good notes: uses exact quotes when possible, uses pseudonyms for
Observations confidentiality, describes activities in the order they occur, provides descriptions without
inferring meaning, includes relevant background info, separates thoughts and assumptions
from what is observed, records time + date + place

 Approaches:
o Direct (watching, listening) vs. indirect (read or heard from 3rd party)

o Structured vs unstructured – what is observed, how will it be recorded, and

interpreted VS monitors all aspects that might seem relevant (problem not so precise)

o Revealed vs unrevealed – participants are aware or not, ethical concerns with

concealment

o Natural vs contrived settings – no special elements VS different arrangements

/artificial

o Participant vs non-participant – ‘living’ with the participants and sharing

experiences VS only spectator

15
  Issues:
o Objectivity/subjectivity (everyone perceives observations differently, might affect
data collection and interpretation)
o Observational reactivity (when being observed you behave differently – people react
differently when they know they are being observed; plus the researcher may also go
native)

Goal: used when the researcher is interested in establishing cause-and-effect relationships.

 To establish cause and effect we need:

o The Independent Variable and Dependent Variable should covary
o The Independent Variable should precede the Dependent Variable
o No other factors should be a possible cause of the change in the Dependent Variable
Experiments o A logical explanation is needed about why the Independent Variable affects the
Dependent Variable

 Control of potential alternatives explanations. How to achieve control in experiments?

o Try to eliminate other factors

o Hold factors constant
o Matching / balancing groups (e.g.: gender, experience)
o Randomization (useful when not all critical factors are known)
o Use control groups

 Manipulation of the independent variable to assess its impact on the dependent variable.

Experiments in the lab

 Internal validity – the confidence we place in the cause-and-effect relationship, to what

extent does the research design permit us to say that the independent variable A causes a
change in the dependent variable B?
 External validity – to what extent are the results found in the lab setting transferrable or
generalizability to actual organizational or field settings?

Experiments in the field – an experiment done in the natural environment in which work (or
life) goes on as usual, but treatments are given to one or more groups.

 Higher external validity but lower internal validity compared to in the lab
 Controlled vs real world settings

Simulations – an experiment conducted in a specially created setting that very closely represents

16
 the natural environment in which activities are usually carried out

 Artificial but close to reality

Intervening factors:

 History effects (influence of external events or occurrences on experiment)

 Maturation effects (natural changes in participants over time / biological & psychological
processes)
 Testing effects (always include a pre-test)
 Selection bias effects (selection of participants – improper or unmatched)
 Mortality effects (attrition of participants in the groups or as the experiment progresses)

Types of design:

 Quasi-experimental designs
o Expose an experimental group to a treatment and measure its effects
o No comparison between groups OR
o no record of the dependent variable before the treatment

 True experimental designs

o Include both the treatment and control groups and record information both before and
after the experimental group is exposed to the treatment.
o Use of double-blind designs to reduce bias (experimenter does not know who is
assigned to what condition)

Ethical considerations:

 Putting pressure on individuals to participate in experiments through coercion or

applying social pressure.
 Giving menial tasks and asking demeaning questions that diminish participants’ self-
respect.
 Deceiving subjects by deliberately misleading them as to the true purpose of the
research.
 Exposing participants to physical or mental stress.
 Not allowing subjects to withdraw from the research when they want to.
 Using the research results to disadvantage the participants, or for purposes not to their
liking.
 Not explaining the procedures to be followed in the experiment.
 Exposing respondents to hazardous and unsafe environments.
 Not debriefing participants fully and accurately after the experiment is over.

17
  Not preserving the privacy and confidentiality of the information given by the
participants.

Action Research

Beyond data analysis - providing recommendations, follow up actions and controls, and
taking decisions.

Stages

1. Initial statement of problem/issue

2. Exploration of problem to identify possible causes: redefinition of problem
3. Generation of alternative solutions
4. Evaluation of options
5. Suggested course of action

18
19