Research methods in Business studies (4th edition)

Pervez Ghauri, Kjell Grnøhaug

Chapter 2: Research in business studies:

 The basic purpose of research education is to teach student to work systematically(page 9)

 Several studies have revealed that mangers do not know how to use research findings and
therefore cannot utilize the result and conclusion of research. ( page 10)
 If the role of a researcher is that of an observer, then what is the difference between an
observer who can draw conclusions with common sense and a researcher?
o The observations made by researcher should be systematic, arguable and challengeable
o Two different people observing the same object may see two different things based on
their background. It is very important to discuss the object and observer biases. (page
11)
o Stress on relationship between case and effect [systematic and based on logic not
beliefs] (page 13)
o 4 differences between common sense and science

laypersons Scientist
Concept and theories Used loosely Build up theories and test
Accept explanations that fits them for internal and external
with their beliefs consistency
hypotheses Evidence come from own Test assumptions and
hypotheses hypotheses systematically
External influences Do not try to control external Study and control relations
influences
explanations Metaphysical explanations Explanations must be able to
tested and studied
(Page 13)

 ‘originality’ describes studies that create a new dimension to already existing knowledge
(page 14)
 Two ways to reach a conclusion:
1. Induction: draw general conclusion from empirical observation.
Observation findingtheory building.
Goes from assumption to conclusions. (page 15)
2. Deduction: draw conclusion from logical reasoning. Build hypotheses from existing
knowledge
Theoryrest of research process.
Associated with quantitative type of research. (page 15)
 Induction and deduction are not exclusive of each other and induction includes elements of
deduction and vice versa. (page 16)
 Ways of knowing: (the first three could lead to false knowledge)(page 18)
 1. Method of tenacity: (truth we know to be truth)
2. Method of authority: (bible)
3. Method of intuition: (self evident)
4. Method of science: (final conclusion is the same for all, self corrective)
 Why research never stops?
The wheel of research (page 19)
 Research and ethics:
o Researcher have moral responsibility to explain and find answers to their questions
honestly and accurately
o Ethical responsibility of a researcher stars with problem formulation.
o Research should not cause embarrassment or other disadvantages to people who have
provided him with data
o There are differences between whether or not something is legal and whether or not it
is ethical.
 Ethical issues in the researcher-participants relationship: Table 2.1 (page 21)
 10 points to improve researcher-participant relationship. (Page 22)
 Ethical issues arise particularly during data collection stage.
 Factors influencing research: (page23)
o Public interests
o Company interests
o Government rules and regulations
o Researcher’s own interests
o Peer pressure
Chapter 3: The research process:

3.1 The process perspective

 Research topic is usually broader and more general than research problem.
 Research design relates to choice of strategy to collect the data needed to answer the stated
research problem

3.2 levels of research:

 Research has two levels conceptual and measurement.

  Research design may be seen as bridge between activities and the conceptual and empirical
levels
 All research requires activities at the conceptual level.

3.3 Research and knowledge:

 The main aim of research is to produce insight or knowledge.

 Knowledge classified as:
1. Theories
2. Concepts
3. Methods/techniques
4. Facts
 Any research should have an intended contribution (bring or add something new)

3.4 What comes first: theory or research?

 We have two research strategies:

1. Identify concepts, theories and adjust them to problem
2. Identify relevant factors and construct explanations

Knowledge (theory) concepts, methods, facts, techniques

2 1

Observations/ assumptions
Problem

3.5 Some important issues:

 Concepts: is an abstracting representing an object.

o Concepts functions:
1. Foundation of communication
2. Introduce a perspective
3. Means of classification and generalization
4. Components of theories
 Definitions: Have two types: conceptual and operational.
1. Describes concepts (clarification and precision of concepts)
o Use clear terms
o Point out unique
o Stated positively
o not be circular
 2. Sets of procedures that describe the activities to be performed to establish empirically
the existence or degree of existence of what described by concept.
 Theory:
o Set of interrelated concepts, definitions, and propositions that present a systematic
review of specifying relations among variables with purpose of explaining and predicting
phenomena.
o Theories have specific aspect to capture the actual problem, and it is almost impossible
to take everything into account at the same time.

 Methods: Rules and procedures

o Logic
o Rules of communication
o Rules of inter-subjectivity, i.e. outsiders should be able to examine and evaluate
research finding.

Problem  Solution

Method

3.6 Concluding remarks:

 Research methodology: a system of rules and procedures

 Rules of reasoning, inter-subjectivity and rules for communications
Chapter 4 research problems:

4.1 research ideas:

 Any idea or observation that attracts attention can be a starting point for research.
 The purpose of any research is to bring a new insight, however when replicating a previous
research there is an aspect of novelty.
 Source for research ideas:
o The real world.
o Looking for missing holes, an important source of research ideas is the past literature.

4.5 The role of reviewing the past literature:

 The prime purpose of literature review:

o Frame the problem under scrutiny.
o Identify relevant concepts, methods/ techniques and facts.
o Position the study.
 The researcher should be to state the criteria of inclusion of the written literature review.
 Systematic search for relevant contribution.
 Writing up the literature review.
 We should evaluate and critique the literature reviewed.
Chapter 5: research design.

5.1 The design problem:

 Poorly formulated research question will lead to misguided research design.

 Effective research design should be able to give the wanted information within the constraints
put on the researcher, (Time, budget and skill constraint).
 “research designs are master techniques, while statistical analysis are servant techniques”

5.2 problem structure and research design:

Main classes of research design:

Research design Problem structure

Exploratory Unstructured
Descriptive Structured
Causal Structured

5.2.1 Exploratory research:

 The exploratory research is the adequate when the research the problem is badly understood.
 The researcher should be flexible
 To take in account the priory information
 And conducted in most possible way
 Requires skills:
o Collect information
o Ability to observe
o Construct explanation (theorizing)

5.2.2 Descriptive research:

 The problem is structured and well understood.

 Key characteristics of descriptive research are structure, precise rules and procedure.
 May contain more than one variable.

5.2.3 Causal research:

 The problem is structure as well.

 Cause and effect problem, the main task is to isolate the causes and to which extent they result
in effects.
5.3 The problem of the ‘cause’

 In order to be a cause:
o Concomitant covariant is needed. Covariation is measures as correlation coefficient.
o The cause should precede the effect.
o Rule out alternative causes.
5.3.2 The importance of theory
 Possible relationships: (PAGE 59-60)
o XY
o YX
o X↔Y
o X≠Y
 Importance of theory:
o Identify the research problem
o Raising questions
o Identifying relevant factors
o Interpreting observations
o Advancing explanations

5.4 The classic experiment:

Most of business studies are not experimental because we cannot control the organizational behavior.

The outside influence is leveled out through randomization.

5.5 validity threats:

 A key purpose of the experimental design it to isolate the effect of potential cause.
 The experiment is considered as a strong design as it allows:
o Manipulation of cause before and after measurement ( measure the covariance
between the case and effect)
o Determine the time order
o Offer some confidence in ruling out the effect of other explanations ( randomizations)
o Valid knowledge  result are true
 Types of validity:
o Internal validity (make sure that x cause y)
o External validity ( the result can be generalized)
 Several threat of validity:
o History: (specific external event occur the same time which may affect the response).
o Maturation: (process is considered as a function of time).
o Test effect: (test may affect the observed response).
o Selection bias: (when the subjects are not assigned randomly).
5.6 other research designs:

 Constraints to experimental research in business:

o The time interval.
o The effects of properties are often general.
o Establishment of identical groups to compare.
o The time order of events is often difficult to determine.

5.6.1 Cross-sectional designs (correlational research):

 Control can be done by partial correlation analysis.

 In cross-sectional research data (dependant and independent) are gathered at the same time.
 A priory knowledge is often used to establish weak causal ordering as in so called path analysis
and causal modeling or SEM (structural equation modeling)

5.6.2 Time Series:

The research acquires observations over time.

5.7 requirements in research design:

 The research design represents the overall strategy on gathering the information needed to
answer the research problem under scrutiny.
 Requirements of research design:
o What is the research problem?
o How it should be represented?
o The potential of hypotheses?
o What are the requirements should the actual research design satisfy?

5.7.1 Research and choices:

 Important decisions in research design:

o How should the concepts be measured?
o What type of data is needed (primary or secondary)?
o If secondary: what secondary data sources are available?
o If primary: how should the data be gathered (observations or interviewing)?
o How should the samples selected?
o How many should be the sample size?
Chapter 6: Measurements.

6.1 defining measurements:

 The key element of measurements is mapping of some properties.

 Measurements can be defined as rules for assigning numbers to empirical properties.
 A measurement is a rule of specifying the procedure according to which numbers are to be
assigned to objects.
o Rules are most significant components of measurement procedure because they
determine the quality of measurements.
o Poor rules make measurements meaningless.
o Rules tie measurements to some aspect of reality.
o Meaningful measurements it has empirical correspondence with what it intended to
measure.
 Why people disagree on their judgment?
o It is not clarified what aspects should be emphasized.
o The rules followed may vary across observers.

6.1.1 objects, properties and indicators:

 We do not measure objects or phenomena as such, rather than measuring specific properties of
the object or phenomena.
Object/ Phenomena
 To map properties we use indicators. Indicator Properties

6.2 Levels of measurements:

The more powerful scales include the properties possessed by the less powerful ones.

6.2.1 Nominal level.

 The lowest level of measurement.

 Objects that are alike assigned to similar number.

6.2.2 Ordinal level.

 It represents variables in some kind of relation, allowing for rank order.

6.2.3 Interval level.

 The distance between each of the observations is constant.

6.2.4 Ratio scale.

 It also possesses a natural or absolute zero.

6.3 validity and reliability in measurements:

 Measurements contain errors.

 Observed measures affected by:
o Stable characteristic (people response).
o Transient personal factor (mood).
o Situational factors (time pressure) and mechanical factor (mark wrong box).

6.3.1 Validity and reliability:

XO=XT+XS+XR

XO: observed score

XT: true score

XS: systematic bias

XR: random error

 Validity: the observed score is close to or equal to true score.

 Reliability: the stability of measure.
 Valid measures presume reliability (random error is modest).
 Reliable measure does not need to be valid.
 Not reliable measures are not valid for sure.
6.3.2 Multiple indicators:

 In business multiple indicators are often used to capture a given construct.

 Crohnbach’s α: measure the inter-correlations between the various indicators to capture the
underling construct.
 Reflective measurement: the various indicators are reflective of the underlying concept.
 Formative measurement: elements supposed to map the underlying construct.

6.3.3 Construct validity:

 The extent to which an operationalization measures the concept which it purports to measure.
o Face validity (reasonable measure).
o Convergent validity (multiple measures/ methods yield similar results).
o Divergent validity (construct is distinguishable from another construct).

6.3.4 Other forms of validity:

 Internal validity: The extents to which we can infer that causal relationship exist between two
variables.
o Correlation coefficient dose not tell us anything about direction
o Correlation coefficient does not indicate a causal relationship
 Statistical conclusion validity: is a prerequisite for making inferences about causal relationship.
o Effect size
o Sample size
 External validity: it relate to what extent the finding can be generalized to particular persons,
setting and times, as well as across types of persons, setting and times.
 The lack of construct validity makes the finding meaningless and also destroys the internal and
external validity.

6.4 improving your measurements:

1. Elaborating the conceptual definitions.

2. Operational definitions.
3. Measures should be corrected and refined (multiple measurements, indicators correlate
positively).
4. The measures should be pre-tested.
5. The final measurement instrument is used in the study.
Chapter 7: Data source.

 Quite often some research questions can best be answered by combining information from
secondary and primary data (page94).
 It is recommended to start with secondary data. ‘Do not bypass secondary data , and only when
the secondary data are exhausted or show diminishing returns, proceed to primary data’
(CHIRCHIL;1999;215)

7.1 secondary data:

 Answer the research question or solving some or all research problems.

 Helping in problem formation and/or devising more concrete and focused research questions
 Deciding on the appropriateness of a certain research method or suggesting better methods for
a particular problem.
 Providing benchmarking measures and other finding that can be compared later on with the
result of the study at hand.

7.1.1 Advantages of secondary data:

1. Enormous saving in time and money

2. Help the researcher to better formulate and understand the research problem
3. Boarder the base which scientific conclusion can be drawn
4. Verification process is more rapid
5. Reliability of the information and conclusion is greatly enhanced
6. Provide excellent historical data
7. Helpful in segmentation and sampling of the target group
8. Facilitate cross-cultural/international research.
9. Suitable methods or data to handle a particular research problem
10. Provide a comparison instrument
11. Save time and facilitate handling of our research questions.
12. Inexpensive and relatively easy to access.
13. Help us understand the situation/research field and identify areas of potential concerns that
merit in-depth investigation based on primary data.

7.1.2 Disadvantages of secondary data:

 We should make a list of the terms and concepts on which we need to collect information.
 If secondary data do not fit with specific problem, we should not use it. And it is better to
answer the question partially.
1. May not completely fit our problem.
2. Companies utilize wishful thinking rather than facts.
3. In developing countries different sources reports different values.
7.1.3 Types of secondary data:

 Internal sources
 External sources; Systematic search: list of main concepts and key words for your research
problem.
o Published
o commercial

7.2 Primary Data:

 When secondary data are not available or cannot answer our research question we must collect
the data that are relevant ourselves.
 Several options for collecting primary data:
o Observations
o Experiments
o Surveys
o Interviews
7.2.1 Advantages of primary data:

They are collected for the particular project at hand.

7.2.2 Disadvantages of primary data:

 Can take long time

 Can cost a lot to collect
 Difficult to get access
 We need to be careful in using proper tools, procedure and methods of analysis.
 Research has less degree of control in data collection.
 Researcher is fully dependant on the willingness and ability of respondents.
7.2.3 Types of primary data:

 Status and state of affairs data

 Psychological and life style data
 Attitude and opinion data
 Awareness and knowledge data
 Data on intentions
 Data on motivations
 Data on behavior
Chapter 8: Data collection

 When and which research approach depends upon:

o The type of the research question.
o The control of the researcher on behavioral events.
o The focus on a current as opposed to a historical phenomenon.
o What information is needed?
o How this can be obtained.

8.1 Qualitative VS quantitative methods:

 The best technique and method is depending on the research problem.

 Methods:
o Historical review and analysis
o Surveys
o Field experiment
o Case studies
 Technique: step by step procedure that we follow in order to gather data and analyze them to
find the answers to our research questions. (How to do? What to do? And Why to do?)
o Structured.
o Semi Structured.
o Unstructured.
  The main difference between quantitative and qualitative is not quality but of procedure.
 Quantitative researchers employ measurements
 The difference between quantitative and qualitative methods and approach is not just a
question of quantification, but also a reflection of different perspective on knowledge and
research objectives.
 It is quite possible to quantify qualitative data.
 Qualitative and quantitative methods are therefore not mutually exclusive.
 Qualitative and quantitative methods can be combined and used in the same study.
 Qualitative research is a mixture of rational, explorative and intuitive where the skill and
experience of the researcher play an important role in the analysis of data.

8.2 When to use qualitative methods:

 No method is entirely qualitative or quantitative.

 Qualitative methods are flexible and unstructured; employ limited number of observations to
explain different aspect of our problem area.
 In such research problems: Uncovering a person’s experience or behavior or Uncover and
understand a phenomenon about which little is known (inductive and explorative research).
o Social.
o Behavior.
o Function (organizations, groups and individuals).
 The three major components of qualitative research:
o Data (collected by interviews).
o Interpretative or analytical procedure.
o Report (thesis or project).
 To support qualitative data : (Miles, 1979:117 ; page 106)
 At the first level the problem is unstructured nature and qualitative methods are suitable. At
second level quantitative methods are most useful, as we want to test different hypotheses. At
third level combination of both is used at this level.
  Low numbers of observations is justified because we want to do in-depth studies or provide
thick description, which not possible with numerous numbers.

8.2.1 Historical review:

 To describe the past to understand present and plan for future.

 The main problem is to trust the human memory, which records selective parts of reality.
 We should cross-check one written source with another, or interview, or two interviews with
each other.

8.2.2 Focus groups:

 Cheap and convenient way of gathering information from several respondents in a short time.
 We should be aware of the influence the group itself will have on the discussion and information
that is exchanged.
 influence factors are:
o Group size
o Composition
o Personality of people
o The role they are asked to play
o Chemistry between the interviewer and the group.

8.2.3 Case studies:

What is case study?

 Case study is descriptive of management situation.

 Case study is useful for theory development and testing.
 It is associated with descriptive or exploratory research (without being restricted to these
areas).
 It used when we need to study a phenomenon in its natural setting (and it is difficult outside)
and also when the concepts and variables are difficult to quantify.

When to use a case study?

 When the researcher has a little control over events and when the focus is on a current
phenomenon in real life context.
 Case study are often:
o Explanatory
o Exploratory
o Descriptive
 When how and why questions are asked, a case study method as a research strategy is favored.
  (What? any of the 5 research strategies. How many? And How much? survey or archival
strategies)
 Single organization and we want to identify factors involved in some aspects or behavior of an
organization or smaller unit.
 Comparative case study: study number of organizations with regards to set of variables we have
identified or assumed.
o Compare the phenomenon in a systemic way.
o To generalize our finding to other organization from the same type.
 Review of existing historical material and records plus interviews (differ from historical review
by the possibility of direct observation and interaction)
 The case study method is not synonymous with qualitative research; it may very well involve
quantitative methods or even be entirely quantitative.

Preparing for a case study

 Theory/data/theory revision:
o Drift: Learn the area of research, concept and terminology.
o Design: choice of strategy to collect data needed to answer the research questions. And
developing tentative explanations of the observations so far.
o Prediction: middle to late stage of the project, case constriction and analysis;
understanding of factors affecting the case.
o Disconfirmation: testing analysis of the result
 Practical approach:
o What are the skills needed?
o What types of priori assumptions do we have?
o How do we select the cases?
o How many cases shall we include in our study?
o How shall we conduct the case study?

How to select the cases

 As other data collection methods:

o Decide the target population
o Assess the accessible population
o Select the cases for the study
 Time and budget are very important.
 In big organizations select the right department or person.
 There are no upper or lower limits for number of cases. Often one case is enough.
Different types of case study design

 Single case:
o Particular case is critical, and to test an established theory.
o Extreme or unique case.
o Revelatory.
o Pilot study or an exploratory study.
 Multiple case:
o In studies not involve critical, rare or revelatory cases.
 Single unit of analysis:
o Inductive approach with specific explanation.
 Multiple units of analysis:
o Inductive but looking for general explanations.
 We can use deductive approach with case study strategy if we are at an early stage to develop
our hypnoses or propositions.
 Deductive approach and looking for generalizations case study is less recommended method.
8.2 Observation:
Chapter 9: sampling ion empirical research.

9.1 Why take samples?

 Two reasons for taking a sample instead of including all units: (page138)
o The cost of including all units will often be prohibitive
o The time needed to do so will often be long

9.2 sampling procedure:

 Sampling procedure can be divided into:

o Probability: non zero chance of being included in the sample.
o Non-probability: (samples are not representative), may give misleading results. Give no
basis for evaluating the size of the sampling variation and error of estimation.
 Convenience: accidental samples.
 Judgment: (we think that sample is representative of the population).
 Quota: choose certain sub groups to represent population.

9.3 probability samples:

 Allow for estimation of the amount of ‘sampling error’.

9.3.1 Simple random sampling

 Simple random: all units in the population have the same chance of being included.
 Questions to be considered:
1. What is the basic unit to be examined?
2. How should the population delineated?
3. What variables or parameters are of interest? (means, variance, ratios, and regression
coefficients)
4. How should the sample be drawn?
5. How many units should be included? (Sample size).
 Advantage: it is easy and easy to understand and apply.
 Drawbacks:
o A complete frame (list of all units in the whole population).
o The cost of obtaining the sample could be very high (interviews).
o The standard errors of estimators can be high.
 Sampling with probability: probability of inclusion in the sample proportional to the value of
variable.

9.3.2 Systematic sampling

 Selecting every nth unit after random start.

 Advantages:
o It is simple
 o Frame is not always needed.
 Disadvantages:
o Danger of hidden periodicities.

9.3.3 Stratified sampling

 The idea about stratified sampling that each subset (strata) gets better representation.
 This method is very important especially when proportions or what we want to estimate are
very different in different strata.
 Stratified sample is a probability sample:
o The parent population is divided into mutually exclusive and exhaustive subsets.
o A simple random sample of units is chosen independently from each subset.
 Variability and standard errors are may be reduced.
 Stratum: part or subdivision of the population.
 Proportional allocation: the proportion of the units included in the sample is the same for each
stratum.
 Advantages:
o Higher precision with the same sample size.
o Gives separate result for each stratum.
o Simplifies data collection.
 Disadvantages:
o Completer frame is needed
o Additional information is needed (SD may be needed for each stratum)

9.3.4 Cluster sampling

 Population is divided into mutually exhaustive subsets.

 One stage cluster sampling: If the researcher examines all units in the cluster.
 Two stage cluster sampling: if the researcher select a sample of units of each cluster.
 Advantages:
o We do not need a complete frame of the secondary sampling units. We do need a frame
of the clusters.
o Concentration of units to be interviewed.
 Disadvantages:
o If there is a large variety between clusters the method may yield poor precision.

The Difference between Strata and Clusters

 Although strata and clusters are both non-overlapping subsets of the population, they differ in
several ways:
o All strata are represented in the sample; but only subsets of clusters are in the sample.
o With stratified sampling, the best survey results occur when elements within strata are
internally homogeneous. However, with cluster sampling, the best results occur when
 elements within clusters are internally heterogeneous (each cluster is small scale model
of the actual population).

9.4 determine sample size

Non-response

 The reliability of probability sample can be improved with the sample size.
 Due to non response the effective sample size is reduced.
 The non responses are different from those who respond, so there is no guarantee that the
sample is representative of the population.
 Non response and other non sampling errors are often more threatening than statistical errors.

9.5 sampling in qualitative research

 Sampling is primarily associated with quantitative research to reach a statistical valid conclusion.
 In qualitative research it is associated with the aim to understand, gain insight and create
explanations.
 Sequential sampling: is continuing to add observations until a final conclusion is arrived.

Theoretical sampling

 In any study variability is needed.

 Purposeful, non-probability samples are often applied in qualitative research.
 Stats means business

John Buglear

Second edition

Chapter 2: Presenting Data