Research Design

MODULE IV
 Why Research Design
Given certain preconditions, the researcher has multiple approaches to study
the same problem (Hitt et al., 1998).
For the same research question, both qualitative and quantitative approach
could be taken (Bartunek et al., 1993) for example, to establish the human
development status of a country, we can look at the quality of life (qualitative)
that people enjoy or look at certain quantifiable parameters like longevity,
literacy and purchasing power parity (quantitative)
Specifying the research question – what to study
Research design specifies the method of achieving the stated research
objectives in the best possible manner – how to study
Different paradigms will guide the selection of the gamut of techniques
available. These differences in approach have led to varying definitions of what
constitutes a research design.

2
Definition & Nature
A Research Design is a logical and systematic plan prepared for directing a
research study. The essential requirement of the design is to provide a
framework and direction to the investigation in the most efficient manner

A traditional research design is a blueprint or detailed plan for how a

research study is to be completed—operationalizing variables so they
can be measured, selecting a sample of interest to study, collecting data
to be used as a basis for testing hypotheses, and analysing the results.
It constitutes the blueprint for the collection, measurement and analysis
of the data.
It is a series of guideposts to keep one going in the right direction.
This is the tentative plan and deviation from this design is quite normal
in research depending upon the availability of data, methods of
sampling etc.
It expresses both the structure of the research problem and the plan of
investigation
3
Research Design: meaning
Formulated design must ensure three basic tenets:
(a) Convert the research question and the stated assumptions/hypotheses
into operational variables that can be measured.
(b) Specify the process that would be followed to complete the above
task, as efficiently and economically as possible.
(c) Specify the ‘control mechanism(s)’ that would be used to ensure that
the effect of other variables that could impact the outcome of the study
have been controlled.

4
Components of a Research Design
The design should include:
A clear statement of the problem
Procedures & techniques to be used for collecting
information (specification of scaling procedures, making questionnaire,
form of data collection etc.)

Population to be studied (sampling process and size)

Plan of the methods to be used in processing and

analysing data

Research designs vary with the types of research.

5
Approaches
positivistic—the method of enquiry would necessarily be scientific and
empirical. Subsequently, this would require a statistical method of
analysis (Ackroyd, 1996).
constructivists - argue for methods that are richer and more applicable to
the social sciences, unlike the more pedantic experimental approach.
Qualitative is a more definitive choice here than the quantitative
(Atkinson and Hammersley, 1994).
principle of triangulation (Jick, 1979) - advocates the simultaneous or a
sequential use of the qualitative and quantitative methods of
investigation. The proponents state that when the findings from diverse
methods are collated, the results are richer, more holistic, that improves
the sanctity of the analysis.

6
 Illustration
Research Problem:
To investigate the consumer decision-making process for organic food products and to
segment the market according to the basket size.
On conducting an extensive review of the literature, it was found that
organic consumption is not always a self-driven choice; rather it could be the
seller who might influence the product choice. Thus, a research design was
formulated to study the organic consumer’s decision stages.
However, once the design is selected and a proposed sampling plan is
developed, the next step required is that the constructs and the variables to be
studied must be operationalized.
On defining the organic consumer, we realized the significance of the
psychographics of the individual—the attitude, interest and opinion—which
were extremely critical. This led to a revision of the research problem:
To investigate the consumer decision-making process for organic food
products and to segment the market—existing and potential—according to
their psychographic profile.

7
 Research Design • Surveys
•Experiments
•Secondary
data analysis
Exploratory Conclusive •Observation

•Secondary Data/ Resource Analysis

(also case study)
•Structured/ Unstructured Observation Descriptive Causal
•Expert Interviews/Opinion Survey
•Focus group interview

Cross-Sectional Longitudinal

Single Multiple

•Cohort Analysis
8
RDs on a continuum

9
Exploratory Studies
The researcher seeks to gain familiarity and/or achieve new insights into the
problem situation. The design is characterised by great flexibility and
versatility. It is simple, loosely structured and flexible
The sample size is not strictly representative and at times it might only
involve unstructured interviews with a couple of subject experts.
For example, a researcher might decide to do an exploratory analysis of the
new channels of distribution being utilized by the marketers to promote and
sell products and services.
To accomplish this, a structured and defined methodology might not be
essential as the basic objective is to understand the new channels. In case the
findings are of interest, the same may lead to a more structured, academic,
basic research or an applied problem where one may want to establish the
efficacy of different channels.
The researcher may resort to tools such as secondary data sources,
unstructured or structured observations, expert interviews and focus group
discussions with the concerned respondent group.

10
RD for Exploratory Studies
• fast and inexpensive way of collecting information – throw
light on the research topic
Secondary • Industry databases – useful
Data/ • Comprehensive case method -intricately designed and reveals a
resource comprehensive and complete presentation of facts, as they
analysis occur, in a single entity.
• This in-depth study is focused on a single unit of analysis

• conducted when no previous information or data is available on

a topic of research. It is formal and structured in general.
Expert
• might be useful when being authenticated or supported by a
Opinion
secondary/primary research or it might be fluid and
Survey unstructured and might require an in-depth interviewing of the
expert

• technique is originally rooted in sociology and is advocated and

Focus used for consumer and motivational research studies.
group • An interview conducted by a trained moderator in an
discussions unstructured way with a small group of respondents.
Discussion may give rise to unexpected findings
11
Secondary Data/ Resources : Illustration
In the Bank research, company reports, trade association reports were
the secondary data used for exploration
Internal and External Data
Advantages and Disadvantages
Alerts: Methodology used, accuracy, updation, content, dependability
Published external data : Directories, Reports, Census data, Govt and
Non-govt data, Policy Documents etc., Computerised Databases
(online and offline)
Syndicated sources – household panels and institutional audits,
industry surveys
Household surveys aim at psychographic and lifestyle studies,
general studies (purchase/consumption behaviour)

12
 Expert Opinion Survey: Illustration
RP: the merit of marketing organic food products in the domestic Indian market
Expert Opinion Survey:
Doctors and dieticians as experts would be able to provide information about the products
and the level to which they would advocate organic food products as a healthier alternative.
Chefs who are experimental and innovative and might look at providing a better value to
the clients. However, this would require evaluating their level of awareness and perspective
on the viability of providing organically prepared dishes.
Pragmatic retailers who are looking at new ways of generating footfalls and conversions
by offering contemporary and futuristic products. Again, awareness about the product, past
experience with selling healthier lifestyle products would need to be probed to gauge their
positive or negative reactions to the new marketing initiatives.
Caution:
Supporting it with some secondary data or subsequently validating the presumptions
through a primary research is recommended.
Another aspect: no expert, no matter how vast and significant his experience is, can be
solely relied upon to arrive at any conclusions. It is also advisable to quiz different expert
sources.

13
Focus Group Discussion
In a typical focus group, there is a carefully selected small set of individuals
representative of the larger respondent population under study. It is called a focus
group as the selected members discuss the concerned topic for the duration of 90
minutes to, sometimes, two hours.
Usually the group comprises six to ten individuals. The number thus stated is
because less than 6 would not be able to throw enough perspectives for the
discussion and there might emerge a one-sided or a skewed discussion on the topic.
On the other hand, more than 10 might lead to more confusion rather than any
fruitful discussion and that would be unwieldy to manage.
Discussions are carried out in neutral settings by a trained observer, called the
moderator. The moderator, in most cases, does not participate in the discussion. S/he
manages a relatively non-structured and informal discussion,initiates the process and
then maneuvers it only to the desired information needs.
Sometimes, there is more than one observer to record the verbal and non-verbal
content of the discussion.
The conduction and recording of the dialogue requires considerable skill and
behavioural understanding and the management of group dynamics

14
FGD: Illustration
In the organic food product study, the focus group discussions were
carried out with the typical consumers/buyers of grocery products.
The objective was to establish the level of awareness about health
hazards, environmental concerns and awareness of organic food
products.
A series of such focus group discussions carried out across four metros
—Delhi, Mumbai, Bengaluru and Hyderabad
FGDs revealed that even though the new age consumer was concerned
about health, the awareness about organic products was extremely low
to non-existent.

15
Other techniques for Exploratory
Research
Depth Interview: It’s an unstructured personal interview by a skilled
interviewer to make note of underlying attitudes on an issue (discussion on
sensitive topics)
Delphi method: The Delphi method is a technique used in group decision-
making and some forms of qualitative research. It involves gathering a
panel of experts, having them complete a survey or questionnaire
individually, and sharing these anonymised answers within the panel to
allow for feedback and debate.
Observation: direct/ indirect; can be qualitative observation or quantitative
observation; to observe a person and draw the finding from their reaction to
certain parameters. There is no direct interaction with the subject.
◦Participant observation
◦Film, photographs
◦Projective techniques
◦Psychological testing

16
Two-tiered RD
Combination of qualitative and quantitative research – simultaneous or
sequential
Initial stage might be an exploratory design
Then based on that one may go for a more formal and structured
research design

17
Caselet
Fisher Price: Marketing Research Is More than Child's Play
Fisher Price Rescue Heroes

18
Descriptive Research
Design

19
 Descriptive Study : nature

•Chalk out a detailed sketch or profile of the respondent population being

studied. This might require a structured primary collation of the
information to understand the concerned population.
For example, a marketer to design his advertising and sales promotion
campaign for high-end watches, would require a profile of the population
that buy high-end luxury products. Thus a descriptive study, which
generates data on the who, what, when, where, why and how of luxury
accessory brand purchase would be the design necessary to fulfil the
research objectives.
• There might be a temporal component to this design, that is, the
description might be in a stagnant time period or be stretched across
collecting the relevant information in different stages in a stipulated time
period.
20
 Descriptive Study : nature
The studies are also carried out to measure the simultaneous
occurrence of certain phenomena or variables. For example, to
establish the relationship (if any) between market shocks and
investment behaviour we have to go for a descriptive research to
establish the correlation between the two variables under study.

Objective of descriptive research is description of an issue, usually market

characteristics or functions

◦ Descriptions of population characteristics

◦ Estimates of frequency of characteristics
◦ Discovery of associations among variables

21
 Descriptive RD
Cross sectional:
The study involves a slice of the population just as in scientific experiments
one takes a cross-section of the leaf
For example, What people think about going on a foreign trip now, when the
COVID is yet to be over (what does people mean? Whom are you studying?)
Longitudinal:
A single sample of the identified population that is studied over a stretched
period of time is termed as a longitudinal study design.
Also called time series design
For example, What people think about the economy in last 1 year (same
sample, repeated observation every quarter!)

22
Cross Sectional RD
The cross-sectional study is carried out at a single moment in time and
thus the applicability is most relevant for a specific period
These studies are carried out on a section of respondents from the
population units under study (e.g., organizational employees, voters,
consumers, industry sectors).

Example: A Danish ice cream company wanted to find out how to

target the Indian consumer to indulge in high-end ice creams. Thus, they
outsourced to a local market research firm to find out the dessert
consumption habits of an upper class, metro Indian consumer. The study
was conducted during March–May 2008 on 1,000 Indian metro
consumers in the upper income bracket.

23
 Example of Cross Sectional
Study
The survey showed that most Indians prefer to eat their specific regional
concoctions at home. But, outside, they love experimenting and generally look at
exotic, foreign desserts or if lost for choice, opt for an ice cream, especially in
summer.
Highlights:
• 92.6% of the sample stated ice cream as the first plus the second choice (81%
first choice). Regional brands were the popular choice of most consumers.
• The recall of foreign brands was, however, only 15%.
• The recall of foreign brands amongst globetrotters (who had made at least five
trips to a foreign country in the last two years) was 39 per cent.
• 76% believed, what you eat and 85% believed where you eat, are influenced by
the social class you belong to.
• 83% usually eat out once every fortnight, 72% once every weekend. 64% eat an
ice cream outside at least once a week. 61.5% were willing to experiment with
exotic desserts, even if they were exorbitantly priced.

24
Example……. continued
The ice cream company concluded from the findings that
the market, at least in the metros, was ready.
However, it was a niche segment and a better audience
base could be found amongst the savvy urban Indian
traveller.
Though the ice cream was healthy and natural, it would
have to take a lifestyle positioning in order to melt the
Indian heart.

25
 Single and Multiple Cross-Sec
Study
There are situations where the population being studied
is not of a homogeneous nature and it is then essential to
study the sub-segments independently.
This variation of the design is termed as multiple cross-
sectional studies.
Usually this multi-sample analysis is carried out at the
same moment in time.
For example, To know the perspective of changes in
tariff on gold, we survey both consumers and producers.
However, there might be instances when the data is
obtained from different samples at different time intervals
and then they are compared.
26
Cohort Analysis
Cohort analysis is such a cross-sectional survey conducted on
different sample groups at different time intervals.
Cohorts are essentially groups of people who share a time zone
or have experienced an event that took place at a particular time
period.
For example, if we would like to know how reading habits have
changed over last 20 years among young people, we need to take
different samples 20 years apart!
The technique is especially useful in predicting election results,
cohorts of males–females, different religious sects, urban–rural or
region-wise cohorts are studied by leading opinion poll experts
like Nielsen, Gallup and others.

27
Longitudinal Study
A panel of consumers specifically chosen to study their grocery
purchase pattern is an example of a longitudinal design.
The study involves the selection of a representative panel, or a group of
individuals that typically represent the population under study.
The second feature involves the repeated measurement of the group
over fixed intervals of time.
A distinguishing and mandatory feature of the design is that once the
sample is selected, it needs to stay constant over the period of the study,
i.e. the number of panel members has to be the same. Thus, in case a
panel member due to some reason leaves the panel, it is critical to
replace him/her with a representative member from the population under
study.

28
Illustration (Longitudinal Design)
The customer portfolio management division of a large private bank wanted to
study the investment behaviour of bank customers in government instruments,
mutual funds and securities, bullion and fixed deposits.
This analysis was done for every quarter for a period of five years. The survey
was done on a different but stock sample of 1,000 bank customers for each
quarter. - THIS IS CALLED AN OMNIBUS PANEL
 Two conclusions pertaining to the researcher’s attitude emerged. First,
government instruments were the most popular option, with approximately 45
per cent customers. Second, the overall percentage of the division amongst the
other three options is more or less stable over time.

29
Illustration….. continued
Another option that the bank had was to form a panel of the regular
customers and assess their periodic investments in these instruments;
here the same group of people would be interviewed in the five-year
period.
The findings and conclusions obtained here would be slightly different,
in case the sample remained the same.
Such a panel study, in addition to indicating an overall investment
behaviour, would have made it possible to monitor the options balanced
between each other by the same group over time, and also how overall
the quarter still showed a uniform pattern.
This data will be available only if the customers studied remain constant
at each data collection phase.
THIS IS AN EXAMPLE OF TRUE PANEL
30
True Panels
Advantages
It has a more committed sample group that is likely to tolerate extended or long
data collecting sessions.
Profile information is a one time task. Thus, a useful respondent time can be
spent on collecting some research-specific information.

Disadvantages
Getting a committed group of people for the entire study period is difficult.
Also, there is an element of mortality and attrition where the members of the
panel might leave midway and the replaced new recruits might be vastly
different and could skew the results.
highly structured study situation which might be responsible for a consistent
and structured behaviour, which might not be the case in the real or field
conditions.
To deal with this, the research agencies try to make certain that people behave
normally and do not demonstrate exaggerated or artificial behaviour.
After a certain period of time, the panel members are changed so that new
perspectives can be obtained.

31
 EXERCISE 1
You are a research executive with a university offering a number of postgraduate courses
like M Com, MCA and MBA. Though any kind of educational qualification enhances
one’s personality, you believe that the two-year MBA programme offered by the
university has a slow and steady impact on the personality development (especially in
terms of introversion/extroversion) of the students.
What is the recommended research design? Justify your selection.
What would be the variables, hypotheses and the population under study?
Longitudinal under descriptive research design
Variables: participation in extra-curricular activities, leadership, academics, asking
questions in class, confidence, self-assessment, goals, soft skill courses, involvement in
club activities/ fests, making friends, getting along in groups, performance in group
activities etc.
Hypothesis:
H1: Going through MBA course makes one more confident
H2: Doing an MBA increases extroversion of students
Population to be studied: MBA students – while entering the course and while passing
out
32
Case Studies
Case Study 1: Keep Your City Clean (Chawla and Sondhi)
Case Study 2: Danish International (Chawla and Sondhi)

33
Research Proposal
What is it and how does it look like?

34
Ref
Chawla and Sondhi, Chapter 3

35