Research Design

Shailendra Singh, Ph.D.

IIM Lucknow
 Overview
• Why Management Uses the Scientific
Approach
• Why You Should Understand Research
Design
 Why Management Uses the
Scientific Approach

• The Characteristics of Science

• The Characteristics of Management
 Why You Should Understand
Research Design
1. To Understand Psychology
2. To Read Research
3. To Evaluate Research
4. To Protect Yourself from Quacks
5. To Be a Better Thinker
6. To Be “Scientifically Literate”
7. To Increase Your Marketability
8. To Do Your Own Research
 Research Design

• The research design is the master plan specifying

the methods and procedures for collecting and
analyzing the needed information.
 Recap
Step 1: Identify and define the Problem or
Opportunity

Step 2: Define the Research Problem

Management Problem Research

Problems
• Focus on symptoms
• Focus on causes
• Action oriented • Data oriented
Problem Formulation and Approach

• Understanding the managerial decision

problem/opportunity
• Translate into research objective(s) (what
information is needed)
• Pose the right research questions
• State questions in terms of hypotheses
• Know what information is needed
 Generating Research Ideas
• From common sense
• From previous research
 Generating Ideas from Common
Sense
• Question everything, • Questions to ask about a
including old sayings: phenomenon:
– True? – Who/When/Why/How
– When is it not true? – What are its effects (short-
– Why is it true? term, long-term, good, bad)?
• Attack practical problems
 How to Get Research Ideas from
Previous Research
• Repeat studies • Improve the study’s
• Do a study suggested by a construct validity
journal article’s author(s) • Look for practical
• Improve the study’s implications of the research
external validity • Try to reconcile studies that
• Improve the study’s produce conflicting results
internal validity
 Generating Research Ideas from
Previous Research: Conclusions
• Easy way to get started in research
• Likely to produce research that is relevant
rather than trivial
• Consider improving study by improving its
internal, external, or construct validity
• Consider extending study by looking at
practical implications, moderator variables,
and mediating variables
 Refine It: 10 Time-Tested Tips
• Don’t be afraid to be • Look for moderator
wrong variables
• Don’t be afraid to deal • Look for mediating
with constructs variables
• Don’t avoid theory • Be more specific
• Manipulate variables (functional
• Look for other effects relationships)
• Reverse cause-effect • Look at components
 Hypothesis Checklist
• Can it be proven wrong?
– Specific prediction? • Are there implications of
– Operational your hypothesis for
definitions? theory, previous research,
• Can it be supported? or a real-world concerns?
• Is it practical and ethical
– It can’t be a null
to test your hypothesis?
hypothesis
• Do you have a rationale
for your prediction?
 Overview
Questions about Applying Techniques from
Physical Sciences to Management
• Validity Questions
– Internal Validity Questions
– Construct Validity Questions
– External Validity Questions
• Ethical Questions: Should it be conducted?
Internal Validity Questions:

Did the Treatment Cause a Change in

Behavior?
Construct Validity Questions:

Can We Make the Leap from the

Physical World to the Mental
World?
3 Follow-up Questions*
 3 Key Construct Validity
Questions
1. What does the measure really measure?
2. What does the manipulation really
manipulate?
3. Is the participant’s behavior genuine or an
act?
 External Validity Questions: Can the
Results be Generalized?
• Can the results be generalized to other
participants?
• Can the results be generalized to other
settings?
 The Research Process

MANAGEMENT PROBLEM

RESEARCH PROBLEM

RESEARCH DESIGN
Today’s
•Exploratory, Descriptive, Causal
Focus •Primary/Secondary data

DATA COLLECTION

DATA ANALYSIS

RESEARCH REPORT
Step 3: Specify the Research Design
 What type of research is going to be used and the source(s) of the data.

 For example, deciding between exploratory, experimental, or

descriptive research
 Research Design
• Definition
– A framework or blueprint for conducting the marketing
research project
• Components
– Information needed
– Data collection methods
– Measurement and scaling procedures
– Sampling process and sample size
– Data analysis procedures
 A Broad Classification of
Marketing Research Designs
Research Design

Exploratory Conclusive
Research Research
Design Design

Secondary Qualitative Descriptive Causal

Research Research Research Research
 Basic Research Objectives and Research
Design

Research Objective Appropriate Design

To gain background information, to define terms, to clarify Exploratory

problems and develop hypotheses, to establish
research priorities, to develop questions to be
answered
To describe and measure marketing phenomena at a point Descriptive
in time
To determine causality, test hypotheses, to make “if-then” Causal
statements, to answer questions
 Exploratory Research
What are its characteristics?
Research is flexible and unstructured
Findings tentative
Used as building block for more research
e.g. pilot surveys, secondary data, focus groups
Done on a small non-representative sample

Why is it used?
When looking for insights into the problem
To help define hypotheses (but does not test) and key variables
To identify alternative courses of action
 When Information needs vague
 For establishing priorities for further research
 Exploratory Research Cont’d
What Methods Are Used?
 secondary data
 qualitative research
 focus groups
 case studies

When is it done?
 Generally initial research conducted to
clarify and define the nature of a problem

What is it’s main limitation?

Does not provide conclusive evidence - subsequent research
expected
 Conclusive Research
Characteristics
 inflexible, versatile
 Results conclusive
 Research formal and structured
 Used when information needs clearly defined

Why is it Used?
 to provide decision maker with the information needed to
make sound decision.
 Testing hypotheses and insights and examining relationships
 Provides estimates of population parameters
 Conclusive
Research
Design

Descriptive Causal
Research Research
 Descriptive Research
What is the Objective?
To describe something, e.g. Target market population
characteristics, proportion of users, predict future demand,

What are the characteristics of descriptive research?

 More rigid than exploratory

 Problem well understood
 Tests specific hypotheses
 Formal and structured
 Large representative samples
 Provides a snapshot of the market environment
 Descriptive Research
What Methods are Used?
• Surveys
• panels
• scanner data
When Used?
• Often a follow-up to exploratory research
 Examples:

 Market segmentation studies, i.e., describe characteristics

of various groups, size of market, buying power of consumers.

 Determining perceptions of company brand or product

characteristics
 Price and promotion elasticity studies

 Product usage studies that describe consumption patterns

 Sales potential studies for particular geographic region or

population segment,
Advertising studies that describe media consumption habits
and audience profiles for specific television programs and
magazines
“I keep six honest serving
men, (they taught me all I
knew), their names are :what
and why, and when, and
how, and where and who”
--Rudyard Kipling
WestJet wants to know whether it should expand its flight
offerings into the European Market. In designing a
market research project to provide WestJet with
information to help them decide, answer the following
six questions.
1. Who should be considered?
2. Where should the respondents be contacted to obtain the right
information?
3. When should the information be obtained from respondents
4. What information should be obtained?
5. Why are we obtaining the information from the respondents
6. How are we going to obtain information
 Descriptive Research Designs
• Two types of designs
– Cross-sectional designs
• Involves collection of information from sample of
respondents only once
• Could have a single cross-sectional design (only
one sample) or multiple cross-sectional design
(many samples of respondents)
• Most popular design in marketing research
• Example: Sample surveys
 Descriptive Research Designs
(contd.)
• Longitudinal Designs
– A fixed sample of the population is measured
repeatedly, i.e., same respondents studied
over time
– Large amounts of data can be collected but
can be expensive
– Useful for tracking changes in consumer
attitudes and behavior over time
– Example: Diary panel data
 Causal Research
What is the objective of causal research?
 To obtain information regarding cause and effect
relationships
Characteristics?
Independent variable manipulated in a relatively
controlled environment
Main method is experiment
Used to understand which variables are causes
(independent variables), and which variables are the effects
(dependent variables)
 Experiments

• An experiment is defined as manipulating (changing

values/situations) one or more independent variables to
see how the dependent variable(s) is/are affected, while
also controlling the affects of additional extraneous
variables.
• Independent variables: those over which the
researcher has control and wishes to manipulate i.e.
package size, ad copy, price.
• Dependent variables: those over which the
researcher has little to no direct control, but has a
strong interest in testing i.e. sales, profit, market
share.
• Extraneous variables: those that may effect a
dependent variable but are not independent variables.
 Experimental Design

• An experimental design is a procedure for

devising an experimental setting such that a
change in the dependent variable may be solely
attributed to a change in an independent variable.
• Symbols of an experimental design:
• O = measurement of a dependent variable
• X = manipulation, or change, of an independent
variable
• R = random assignment of subjects to
experimental and control groups
• E = experimental effect
 Experimental Design

• After-Only Design: X O1
• One-Group, Before-After Design: O1 X O2
• Before-After with Control Group:
• Experimental group: O1 X O2
• Control group: O 3 O4
• Where E = (O2 – O1) – (O4 – O3)
 How Valid Are Experiments?

• An experiment is valid if:

• the observed change in the dependent variable
is, in fact, due to the independent variable
(internal validity)
• if the results of the experiment apply to the
“real world” outside the experimental setting
(external validity)
 Uncertainty Influences The Type Of Research

CAUSAL OR EXPLORATORY
DESCRIPTIVE COMPLETELY CERTAIN ABSOLUTE AMBIGUITY

Exploratory Research Descriptive Research Causal Research

(Unaware of Problem) (Aware of Problem) (Problem Clearly Defined)

“Our sales are declining and “What kind of people are “Will buyers purchase more of
we don’t know why.” buying our product? our products in a new package?
“Would people be interested in Who buys our competitor’s
“Which of two advertising
our new product idea?” product?”
campaigns is more effective?”
“What features do buyers
prefer in our product?”
 Understanding EXPLORATORY or
Poor
of problem?
Problem? SECONDARY
RESEARCH
Good

Need to Yes EXPERIMENT

establish
causality?
No

Objective Yes Needestimates

Need estimates Yes
answers by
Survey
of of prevalence?
prevalence
asking?
No
No

OBSERVATIONAL FOCUS GROUPS

RESEARCH
Step 4: Develop the Data Collection Procedure
• Decide whether the data will be collected through surveys, interviews,
focus groups, etc.

Step 5: Design the Sampling Procedure

• Decide who will be in the sample, how those people are
selected, and the size of the sample.
Step 6: Collect the Data
• go out and collect as determined in steps 4 and 5

Step 7: Process and Analyze the Data

• Code and then analyze the data by testing hypotheses

Step 8: Present the Results

• Create an oral or written presentation of the results of the
study
 DATA SOURCES

Secondary Primary

Internal External
 Secondary Data
Data gathered and recorded by someone else prior to and
for a purpose other than the current project.
Is often:
• Historical
• Already assembled
• Internal to corporation
 Common Research Objectives
for Secondary Data Research
Fact Finding - Identifying consumption patterns
- Tracking trends

Model building - Estimating market potential

- Forecasting sales
- Selecting trade areas and sites

Data Base - Development of Prospect Lists

Marketing - Enhancement of Customer Lists
 Advantages of Secondary Data
 Inexpensive

 Obtained Rapidly

 Needs no access to subjects or respondents (convenient)

 Information is not Otherwise Accessible

 Can Provide Insights into problem during exploratory phase

 Can provide background data on trends etc. which lends credibility to the
report
 Disadvantages of Secondary
Data
 Lack of Availability (e.g. new products; image)

 Uncertain Accuracy

 Data Not Consistent with Needs (not relevant)

 Inappropriate Units of Measurement

 Time Period Inappropriate (Dated)

 Evaluating Secondary Data

Does the data help to

answer questions
set out in the
problem definition?

Applicability Does the data apply to

to project the time period of
objectives interest?

Does the data apply to

the population of
interest?
 Evaluating Secondary Data (continued)
Do the other terms
and variable
classifications
presented apply?
Applicability
to project
objectives Are the units of
measurement
comparable?

If possible, go to the
Accuracy original source of the
of the data data?
 Evaluating Secondary Data (continued)

Is the cost of data

acquisition worth it?

Accuracy
of the data
Is there a possibility
of bias?

Can the accuracy of

data collection be
verified?
 Internal Data
 Accounting information

 Sales information

 Backorders

 Customer complaints
 Data Mining
The automated extraction of hidden predictive
information from large databases
E.g. Blockbuster mines its video rental history database to
recommend rentals to individual customers
 data mining is used to discover patterns and relationships in the
data in order to help make better business decisions.
Data mining can help spot sales trends, develop smarter marketing
campaigns, and accurately predict customer loyalty.
Uses
Market segmentation - Identify the common characteristics of customers who buy
the same products from your company.
Customer churn - Predict which customers are likely to leave your company and go
to a competitor.
Fraud detection - Identify which transactions are most likely to be fraudulent.
Direct marketing - Identify which prospects should be included in a mailing list to
obtain the highest response rate.
Interactive marketing - Predict what each individual accessing a Web site is most
likely interested in seeing.
Market basket analysis - Understand what products or services are commonly
purchased together; e.g., beer and diapers.
Trend analysis - Reveal the difference between a typical customer this month and
last.
 External Data

Created, recorded, or generated by an

entity other than the researcher’s
organization.
 Sources of External Data
 Libraries
 The Internet
 Vendors
 Producers
 Books and periodicals
 Government
 Trade associations
 Newspapers and journals
 Commercial Sources
 Attitude and public opinion research—syndicated
services report the findings of opinion polls

 Consumption and purchase behavior data

 Advertising research—readership and audience data

 Market share data companies like A.C. Nielsen provide

information about sales volume and brand share over time
You have decided to open a new retail store in Calgary that
will sell personal computers and software.
What information do you need to help you determine where
to locate?

What secondary data are available to help you decide where

to locate the store?