RESEARCH DESIGNS

Tuesday, April 16, 2024 1

 Chapter Outlines
 OVERVIEW
 RESEARCH DESIGN: DEFINITION
 Why Research Design is needed?
 Classification of Research Designs
 Exploratory Research Design
 Conclusive research Designs
 Descriptive Research Designs (DRD)
 Methods of Descriptive Research
 Cross-sectional and Longitudinal Designs
 Causal Research desaigns
 Scientific Notion of Causality
 OVERVIEW
 After the initial phase of research process, the
researcher designs the formal research project and
identify the appropriate sources of data for the study.

 We can classify research designs: exploratory and

conclusive. We further classify conclusive research
designs as descriptive or causal and discuss both
types in detail.

 We then consider the differences between the two

types of descriptive designs, cross-sectional and
longitudinal,

Tuesday, April 16, 2024 3

 RESEARCH DESIGN: DEFINITION
 According to Churchill Jr. a research design is simply the
framework or plan to be used as a guide in collecting and
analysing data. It is the blue print that is followed in
completing study. It is, like the architects’ blue print for a
house.
 What should be contained in a research design. may vary
depending on the preferences of a person responsible.
 It is generally agreed that in a research design the details
should be kept as minimum as possible for later
modifications. However, the following should be included in
any types of research.
 Identify the information needed
 Specify the measurement and scaling procedures
 Specify the sampling process and sample size.
 Develop a plan of data analysis.

Tuesday, April 16, 2024 4

 Why Research Design is needed?

 Clarity: Without RD the researcher does not know what to do. It

is extremely desirable that the design be put in writing .
 Relevance: When the researcher is deciding specially what
data are needed.
 Ease in Analysis and Interpretation: Keep the computations
and thinking on the path to solutions and recommendations.
 Economy: Just as the house built without a blue print is likely to
cost more because of midstream alterations in constructions, a
study is likely to cost more without formal design.
 In a nutshell a research design ensures that the study 1) will be
relevant to the problem, and 2) will use economic procedures.
 However, one should not conclude that a design should have a
very rigid framework. There should always be some room for
flexibility.
Tuesday, April 16, 2024 5
 Classification of Research Designs

 Research designs may also be classified as exploratory or

conclusive
 Exploratory research (ER) is used to define the problem more
precisely, identify relevant courses of action, or gain additional
insights. It is flexible and unstructured.
 The sample selected is small and non-representative. The primary
data are qualitative in nature. The findings are tentative. The
insights gained from exploratory research might be verified by
conclusive research.
 Conclusive research is more formal and structured than ER. It is
based on large, representative samples, and the data obtained are
subjected to quantitative analysis.
 The findings are considered to be conclusive. CR designs may be
either descriptive or causal, and descriptive research designs may
be either cross-sectional or longitudinal.

Tuesday, April 16, 2024 6

 DIFFERENCES BETWEEN EXPLORATORY AND CONCLUSIVE
RESEARCH
Exploratory Conclusive

Objective: To provide insights and To test specific hypotheses

understanding. and examine relationships

Characteristi Information needed is defined only Information needed is clearly

c: loosely defined.
Research process is flexible and Research process is formal and
unstructured structured
Sample is small and non- Sample is large and
representative representative.
Analysis of primary data is Data analysis is quantitative.
qualitative. Tentative Conclusive
Findings/
Results: Tentative Conclusive

Outcome: Generally followed by further Findings used as input into

exploratory or conclusive decision making
research

Tuesday, April 16, 2024 7

 Research Designs: Three Caviets
 The crucial tenet of research is that the design of the
investigation should stem from the problem. Each of these
types is appropriate to specific kinds of problems.
 The general principles of each design can be taught.
Whether they are productive in a given situation depends
on how skillfully they are applied.
 These basic designs can be looked as stages in a
continuous process. Exploratory studies are often
considered as the initial step. When researchers begin an
investigation, it stands to reason that they lack a great deal
of knowledge about the problem.
 Later on, if the problem gets crystallize a descriptive or
causal study can be undertaken.

Tuesday, April 16, 2024 8

 CLASSIFICATIONS OF RESEARCH DESIGNS

Research
ResearchDesign
Design

Exploratory Conclusive
Conclusiveresearch
research
Exploratoryresearch
researchdesign
design design
design

Descriptive
Descriptiveresearch
research Causal research design

Cross
Crosssectional
sectional Longitudinal
Longitudinaldesign
design
design
design

Single
SingleCross
Cross Multiple
MultipleCross
Cross True Omnibus
Sectional True Omnibus
Sectionaldesign
design Sectional
Sectionaldesign
design Panel
Panel panel
panel

Tuesday, April 16, 2024 9

 Exploratory Research Design
 The objective in explorative research is to gain insights and ideas of
a problem. Appropriate when the research objectives include:
1. Identifying problems or opportunities,
2. Precise formulation of a vaguely identified problem or opportunity,
3. Gaining perspective regarding the variables operating in a situation
4. Gaining management and researcher perspective of the problem
situation,
5. Gathering information on the problems associated with conclusive
research.
6. Purpose is to formulate hypotheses regarding potential problems
and/or opportunities present in the decision situation.
 Examples of hypotheses developed from exploratory research are
as follows:
 An ad theme emphasizing the "nutritional value" of a food product
will increase brand awareness more than "good flavour" theme.

Tuesday, April 16, 2024 10

 Exploratory Research (Continued)
• In general, ER is appropriate to any problem about which little is
known
• Because ER deals with a decision situation where limited
knowledge exists, ER is characterised by flexibility.
• Formal design is conspicuous by its absence. It rarely involves
structured questionnaires, large samples, and probability
sampling plans.
• Once a new idea or insight is discovered, they may shift their
exploration in that direction.
• Thus, the creativity and ingenuity of the researcher plays a major
role in ER.
• ER can greatly benefit from use of the following methods.
• (1) Literature Search
• (2) Experience Survey
• (3) Analysis of Selected Cases
• (4) Qualitative research techniques

Tuesday, April 16, 2024 11

 Methods Used in ER
• Literature Search: Quickest and cheapest way to discover
hypotheses is in the work of others, through a literature research. The
search would include published work in that area, company records,
earlier studies.
• Emphasis on the discovery of ideas and explanations of phenomenon.
• Experience Survey: (key informant survey,) Taps the expertise of
those familiar with the general subject being investigated. This include
top executives, key managers, consultants, academicians, etc.
• Analysis of Selected Cases: (Analysis of "insight stimulating
examples“). This approach involves the intensive study of selected
cases of the phenomenon under investigation.
• This involves the examination of existing records, observation of the
occurrence of the phenomenon, unstructured interviewing, or some
other approach may be used.).

Tuesday, April 16, 2024 12

 Methods Used in ER (Continued)
• In one study to improve the productivity of the sales force the
researcher studied intensively two or three of the best sales
representatives and two or three of the worst.
• Data were collected on the background and experience of each
representative and then several days were spent making sales calls
with them. As a result a hypotheses was developed.
• It was that checking the stock of retailers and suggesting items on
which they were low were the most important differences between the
successful and the poor sales representatives.
• Qualitative research Techniques:
• It is a major methodology used in ER. Researchers undertake
qualitative research to define the problem or develop an approach.
• Qualitative research uses techniques such as focus groups, depth
interviews, and projective techniques with emphasis on association,
completion, construction, and expressive techniques.

Tuesday, April 16, 2024 13

 Conclusive research Designs

 These are two types: (1) Descriptive research, (2) Causal research.
 Descriptive Research Designs (DRD)
 A great deal of research can be considered DR and is appropriate
when the research objectives include:
 1) Portraying the characteristics of a management phenomenon,
and determining the frequency of occurrence.
 2) Determining the degree to which different variables are
associated,
 3) Making predictions regarding the occurrence of a particular
phenomenon.
 DR is pre-planned and structured. It is based on large
representative samples.
 A DRD requires a clear specification of who, what, when, where,
why, and way (the six Ws) of the research.
 In summary, DR, in contrast to ER, is marked by a clear statement
of the problem, specific hypotheses, and detailed information
needs.

Tuesday, April 16, 2024 14

 Example of Descriptive Research Designs
• A company wants to know the influence of husband and wife (H/W) in
the purchase of its product.
• Some of the possible questions to be answered before the data
collection are: What instrument be used for measuring the (H/W)
influence?
• What stages of the decision making of (H/W) have to be measured?
• Whether the data should be collected from husbands only or both?
• Whether to use structured questionnaire or unstructured.
• The researcher should delay data collection until how the data are to
be analysed has been clearly determined.
• Ideally one would have a set of dummy tables before the collection
process.
• A dummy table is a table that is complete in all respects except the
actual numbers.

Tuesday, April 16, 2024 15

Dummy Table 1: Involvement of Husband-Wife in Refrigerator
Purchase

(N=105, n1 =63, n2 = 42)

Level of Involvement %

Stages In Type of Family W>H H=W H>W 

Purchasing

Single Earning
Family (n1)
Idea Initiation

Dual Earning Family

(n2)

Single Earning
Family (n1)

Information Dual Earning

Collection Family (n2)

Single Earning
Family (n1)

Final Dual Earning

Family (n2)
Decision
Single Earning
Family (n1)
Methods of Descriptive Research

 Secondary data analyzed in a quantitative as

opposed to a qualitative manner
 Surveys
 Panels
 Observational and other data
 We will discuss types of Descriptive Research
Designs: Cross Sectional and Longitudinal
 Cross-sectional Designs
 Involve the collection of information from any given sample of
population elements only once.
 In single cross-sectional designs, there is only one sample of
respondents and information is obtained from this sample only
once.
 In multiple cross-sectional designs, there are two or more
samples of respondents, and information from each sample is
obtained only once. Often, information from different samples is
obtained at different times.
• The CSDs are carried through surveys and also called survey
research design.
• Survey attempts to be representative of some known universe, both
in terms of number of cases included and the manner of their
selection.
 Longitudinal Designs (LDs)
• In LDs a fixed sample (or samples) of population is measured
repeatedly, samples remain the same over time.
• LD provides a series of pictures that give view of the situation and
changes that take place over time.
• Often, the term panel is used interchangeably with the term LD.
• A panel consists of a sample of respondents, that have agreed to
provide information at specified intervals over an extended period.
Syndicated firms maintain panels.
• There are two types of panels. The older type of relies on repeated
measurement of the same variable i.e. each panel member is
measured with respect to the same characteristics at each time.
• A new type of panel called the omnibus panel, a sample of elements is
still selected and maintained, but the information collected from the
members varies. At one time the panel members might be asked to
evaluate alternative advertising copy. At another time it is attitude with
respect to a new product.
• The distinction between the traditional panel and the omnibus panel is
important.
 Example of Longitudinal Design
• A company maintains a panel of 1000 families & it is interested in
determining the impact of recent change in package design for its brand A
and that there are two main competitors B and C and no of smaller
competitors as in the single category Brand D. Let us consider the
performance of our brands at time period to before the change and time
period t1 after the package change.
• Several types of analysis: We could look at the proportion of those in the
panel who bought A in period t0. We could also calculate the proportion of
those who bought A in period t1 .The table above indicates that package
change was successful. A's market share has increased from 20% to 25%.
Further, A seemed to make its gains at the expense of its two major
competitors whose market share decreased.
Brand Purchased At time t0 At time t1
A 200 275
B 300 270
C 350 330
D 150 150
Total 1000 1000

Tuesday, April 16, 2024 20

 RELATIVE ADVANTAGES AND DISADVANTAGES OF LONGITUDINAL
AND CROSS-SECTIONAL DESIGNS

Evaluation Criteria Cross-Sectional Design Longitudinal Design

Detecting change -- +

Large amount of data - +

collection
Accuracy -- +

Representative sampling + --

++Response bias + --

Note: + indicates a relative advantage over the other design; whereas -

indicates a relative disadvantage

Tuesday, April 16, 2024 21

 Causal Research Designs ( CRD)
 CR determines the cause and effect relationships.
 For example, a 5% increase in price of the product will have no
appreciable impact on the quantity demanded by customers.
 DR is not suitable for establishing, causality. Therefore, to
establish causal relationship the experiments are used.
 Causal research is appropriate for understanding which variables
are the cause/s (independent variables) and which variables are the
effect (dependent variables)
 Like DR, causal research requires a planned and structured design.
Causal design, in which the causal or independent variables are
manipulated in a relatively controlled environment. The main
method of causal research is experimentation.
 To understand CRD, one must understand the scientific notion of
causality.
 A COMPARISON OF BASIC RESEARCH
DESIGNS

Exploratory Descriptive Causal

Object Describe market Determine cause

ive: Discovery of ideas
characteristics or and effect
and insights.
functions. relationships.

Chara Marked by the prior Manipulation of one

cterist Flexible, versatile. formulation of or more independent
ics: specific hypotheses. variables.
Expert surveys Secondary data
Metho Pilot surveys Surveys Experiments
ds: Secondary data Panels
Qualitative research Observational and
other data