Research for Marketing Decisions

PGP - Term III (2024-2025)

Mid-term Topics

Conceptualisation of Marketing Research (AMA)

The American Marketing Association (AMA)

redefined Marketing Research as:

The function which links between customers and

managers through INFORMATION

Amazon Influencer Program

Information
Customers Managers
Marketing Research
 Data vs. Information – Mechanism

Data Analysis Information

• Why Data? • Which type of analytics fit • How to use information in

the Data? managerial activities?
• What kind of data?
• Do the data meet the • Do the information solve the
• How to collect Data?
assumptions of analytics? problem or give managers
better businesses?

Usage of Information
Use to identify and solve
marketing
problems/opportunities

Information Monitor marketing

performances

Improve understanding of
marketing as a process
 Definition of Marketing Research (Malhotra and Dash, 2021; Das,
Aaker, Day, Kumar, 2021)

Marketing research is the systematic and objective procedure of

▪ Collection of data
▪ Analysis of data
▪ And use of information

for the purpose of improving decision making related to the

▪ Identification of problems/opportunities in marketing, and

▪ Solution of problems/opportunities in marketing.

Two Activities of MR: Problem identification Vs. Problem

solving research
Marketing Research

Problem Solving Research

Problem
Identification Research
Kellogg’s Nutri-Grain Cereal Bar
Blackberry. In-store promotion

Why Kellogg’s sales of cereal is

declining? - New products for adult
market (problem identification)
 Marketing Research Process

Step 1: Problem identification Mainly,

Qualitative
Research
Step 2: Research objective

Step 3: Research method

Step 4: Data collection

Step 5: Data analysis

Mainly,
Quantitative Step 6: Report writing/Managerial implications
Research

Classification of Marketing Research: Qualitative Vs. Quantitative

Qualitative Quantitative
Objective: To provide insights and To test specific hypotheses and examine
understandings. relationships.

Characteristics: Research process is unstructured. Research process is structured. Sample

Sample size is small and non- size is large and representative. Data
representative. Data analysis is analysis is quantitative.
qualitative.

Techniques: FGI, In-depth interview, Lit review Survey, Experiment

Data Analysis: Nonsatistical Statistical

Findings Tentative Conclusive

/Results:
Generally followed by further Findings used as inputs into decision
Outcomes: quantitative research. making.
IMPORTANCE OF IDENTIFYING THE PROBLEM CORRECTLY

• Only when the marketing research problem has been clearly defined, research
can be designed and conducted properly.

• All the effort, time, and money spent on will be wasted if the problem is mis-
understood and ill-defined.

• Marketing research recommendations may be wrong if the problem definition is

not correct.

• Wrong problem definition/identification can be a cause of brand death.

Steps involved in Problem Identification

Step I: Problem Identification

Symptoms (Managers come across First)

◼ Discussions with Decision Makers

◼ Interviews with Industry Experts MR
◼ Secondary Data review Consultants’
◼ Talking to customers activities
◼ FGI, In-depth interview (Qualitative,
◼ Literature review
generally)
◼ Review of environmental contexts

◼ Ethnography study

Management Decision Problem (MDP) (What a manager NEEDS to do?)

Marketing Research Problem (MRP) (What INFORMATION is needed to answer the MDP?)
Step 1:Problem Identification of Chumbak – A LIVE example

SYMPTOM
Decline in Sales.
Talk to DM, customers, review
of secondary data, FGI

(Potential root cause –

Loyalty)

MANAGEMENT DECISION PROBLEM (MDP)

How to revive Sales through Store loyalty?

MARKETING RESEARCH PROBLEM (MRP)

• Examine the current impacts of factors that
influence Store loyalty of Chumbak.**

Jenish Kamdar (1711022)

**We can have more than one Kucharlapati Ravi Varma (1711181)
MRP also such as “Revisit the Nukala Vijaya Lakshmi (1711194)
current Loyalty Program”. Vishwajeet Karmwar (1711439)
Nicolas Arias (17E5018)

Research Objective

Research Objective: Converting the Business Problem into Research Problem.

Research objective should be very specific and clear (WHAT?)

Marketing Research Problem (MRP)

If the MRP is very (Broad Information)
clearly defined,
it may be considered If needed
as research objective as it is,
otherwise, further qualitative
research is needed to make it
clear for research objective.
Research Objective
(Very specific information)
 How to represent research objective?

Research Hypothesis
statements/questions Modeling

Development
Testing
Mathematical Graphical
-Qualitative research
-Quantitative Research
-Need literature/argument
- Six steps in testing
-Report H1

Literature Review Concept: A MR Perspective

• A literature review is a comprehensive summary of previous market

research on a topic of interest. For example, all scholarly articles exploring
the relationship between price and customer satisfaction.

• The literature review searches scholarly articles, books, MR projects/reports,

and other sources relevant to a particular area of research.

• The literature review acknowledges the work of previous market

researchers, and in so doing, it provides insights and understandings on the
MR problem at hand.

• Literature review gives a theoretical base for the marketing research and
thus enhances insights and understandings on the MR problem.
 Uses of Secondary Data _ A MR Perspective

• Identify the problem/opportunity

• Better define the problem/opportunity
• Develop an approach to the problem/Research Objective
• Formulate an appropriate research hypothesis (for example, by
identifying the key variables)
• Answer certain research questions and test some hypotheses
(Quantitative research – if the data meet the characteristics of
Quantitative Research).

Secondary data can be used in both Qualitative and Quantitative

research, depending on features of the data

A Classification of Qualitative Research Procedures

Qualitative Research
Procedures (Interview)

Direct (Non- Indirect

disguised) (Disguised)

Projective
In-depth Interviews Techniques
Focus Groups

Association Completion Construction Expressive

Techniques Techniques Techniques Techniques
 Focus group interview – Conceptualization

• A focus group is an interview conducted by a trained moderator in a non-

structured and natural manner with a small group of respondents.

• The moderator leads the discussion.

• The main purpose of focus group is to gain insights and understandings by

listening to a group of people from appropriate target market talk about issues
of interest to the researcher.

• Results of focus group interview to be further validated with Quantitative

research.

Focus group interview – Usage

• To identify possible reasons behind symptoms/ Help in identifying problems

• To develop hypotheses about the way consumers think and behave

• To obtain information for structuring questionnaires by uncovering relevant

questions and appropriate response categories.

• To generate and evaluate new ideas for products or product uses

• To develop research models as part of research objective.

• When you need to brainstorm ideas.

• When you want to better understand the complexities of your target market.
 Characteristics of Focus Groups

Group Size 5-10

Group Composition Homogeneous, respondents,

prescreened

Physical Setting Relaxed, informal atmosphere

Time Duration 1-3 hours

Recording Use of audiocassettes and

videotapes

Moderator Observational, interpersonal, and

communication skills of the
moderator

When Focus Group Interview is NOT recommended?

• Sensitive issues (e.g., personal income and saving), embarrassing products (e.g.,
anti-gas medicine, fungal foot powder).

• Less educated respondents (e.g., BOP consumers), introvert respondents.

• When participants are competitors/top business professionals (e.g., B2B context)

• Situations where strong social norms exits (attitude of college students towards
dowry system).

• Very deep insights such as how products attributes lead to achieve customers life
goals.

In such cases, we will be using In-depth interview

 In-depth interview – Conceptualization

• In-depth interview is a direct method (where the researchers disclose the

purpose of the research) of obtaining qualitative data.

• In-depth interviews are an unstructured/semi-structured and conducted on a

one-on-one basis to understand customer’s motivations, beliefs, attitudes and
feelings on a topic.

• An in-depth interview may take from 30 minutes to more than one hour.

In-depth interview – When to use?

• When participants are
competitors (e.g., Car dealers).

• When you’re concerned about

group hierarchy (Boss and
employee).

• Less educated respondents (e.g.,

BOP consumers), introvert
respondents.

• When you’re gathering sensitive

feedback (e.g., political ideology
and product usage).
 Content analysis – Conceptualization

• Content analysis is a Qualitative Data (e.g., FGI, In-depth interview, projective

technique) analysis technique.

• The term “content analysis" is used here to mean the scientific analysis of
content/key words of communications messages.

• Content analysis bridges the gap between qualitative and quantitative research.

Content Analysis Outcomes to

Qualitative
be used for
techniques
Quantitative
(e.g., FGI)
research

Content analysis – Characteristics

Objectivity – Content analysis should be objective. MR consultants should

derive the results based on the actual content of the messages. Two MR
consultants, following identical procedures with the same set of data
independently, should arrive at similar conclusions.

Systematic - The meaning of "system" is that content analysis should lead to

the results required to develop research objectives or hypothesis or purpose
of the study. The findings should have theoretical relevance and be tested
further (quantitative research).

Quantification - This requirement helps us to understand the relative

importance of factors. Quantification can be done with frequency count or
the assignment of numerical values to the content of messages .
 Content analysis – Procedure

Listen the FGI/in-depth Code/number the participants

Interview recording few times. with numerical numbers such as
participants 1,2, 4, 5, … (to avoid
subjectivity, if any).

Go through the written statement Write down the questions of

and/or listen (again, for moderators and answers of each
verification only) the recording few participant derived from the FGI
times and write down the recoding.
factors/key words/statements each
respondent mentioned in a
frequency distribution table.

Projective Technique – ORIGIN and WHY?

• Project Technique (PT) is Originated in clinical psychology. Marketers adopted the

technique from psychology.

• Instead of questioning directly, use 3rd parties: people or brands to enable

subjects to “project”.

• According to PT, when we are asked a straightforward question about a particular

topic, we spend time consciously creating an answers that are socially acceptable or
desirable but are perhaps not the most accurate reflection of our true feelings or
behavior.

• Application Context: When the issues to be addressed are personal, very

sensitive, or subject to strong social norms.
 Projective Technique – ORIGIN and WHY?

• People don’t always share inner feelings

• People are often unaware of their motives, values, attitudes.

• Fear of being considered stupid or irrational

• Reluctant to admit certain types of behaviours

• Allow researchers to tap into consumers’ deep motivations, beliefs, attitudes

and values through indirect approaches.

Advantages of Projective Techniques

• They may elicit responses that subjects would be unwilling or unable to

give if they knew the purpose of the study.

• Helpful when the issues to be addressed are personal, sensitive, or

subject to strong social norms.

• Helpful when underlying motivations, beliefs, and attitudes are operating

at a subconscious level.
 Disadvantages of Projective Techniques

• Suffer from many of the disadvantages of unstructured direct

techniques, but to a greater extent.
• Require highly-trained interviewers.
• Skilled interpreters are also required to analyze the responses.
• There is a serious risk of interpretation bias.
• They tend to be expensive.

Research method: Conceptualization

Research method: Plan/framework for carrying out the

study/research.

• A good research method will ensure that the marketing research

project is conducted effectively and efficiently.
 Components of a Research Method

3.1. Nature of study/research design: Exploratory; and Conclusive

(descriptive, causal).

3.2. Measurement and scaling procedures

3.3. Data collection instrument

3.4. Sample size and sampling

Classification of Nature of study

Research Nature

Exploratory Conclusive
Research Design Research Design

Descriptive Causal
Research Research
A Comparison of exploratory, descriptive and causal studies

Exploratory Descriptive Causal

Objective: Describe market Determine cause
Discovery of ideas
characteristics or and effect
and insights
functions relationships

To describe the
Example: A review of Impacts of
characteristics of
literature to advertisement on
relevant groups, such
identify factors sales
as consumer
influencing
segments.
purchase
behaviour.

Surveys
FGD Experiments
Observation
Methods: Expert surveys
Pilot surveys
Secondary data
Secondary data

Marketing examples of exploratory, descriptive, and causal

studies

Exploratory Descriptive
Causal
Ambiguous problems. Aware of problems Clearly defined problems
“Our sales are declining “What kinds of people are “Will buyers purchase more of
...why.” buying our products? Who our product with
buys our competitors’ a change of our website?”
“Our customers are products?”
leaving…why.” H1: “Price, and Quality are H1: “Would changes in price
related to sales” and quality lead to change in
sales?”

We use the following We use the following

words for descriptive: words for causality:
Related, Associated, Impact, Influence, Lead
Linked. to.
 What could be the nature of your MR project?

• Only exploratory?

• Only Conclusive? – Problem Solving

• Exploratory-cum-conclusive? – Problem identification cum solving

• Exploratory-cum-conclusive-cum-exploratory?

• Conclusive-cum-exploratory?

Look at the Steps I & II of your MR Project to decide the

nature of your MR Project.

Populations and Samples

• A Population (N) is the set of all items or individuals of interest
– Examples: All likely voters in the next election
All Airtel customers
All tax receipts over this year

• A Sample (n) is a subset of the population

– Examples: 1000 voters selected at random for interview
5000 Airtel customers
1000 receipts selected for audit
 Sample Size Determination for Means and Proportions
(formula and example)
Steps Means Proportions

1. Specify the level of precision. D = $5.00 D = p -  = 0.05

2. Specify the confidence level (CL). CL = 95% CL = 95%

3. Determine the z value associated with CL. z value is 1.96 z value is 1.96

4. Determine the standard deviation of the Estimate :  = 55 Estimate :  = 0.64

population.

5. Determine the sample size using the n = 2z2/D2 = 465 n = (1-) z2/D2 = 355
formula for the standard error.

G*Power Software: [Link]

psychologie-und-arbeitspsychologie/[Link]

Formula of sample size determination

For the above formula (refer previous slide), we need to consider the
confidence interval either at 90% or 95% or 99%. This will give us the
Z and D/error value.

The standard deviation (sigma), can be found with a pilot study (n <
30) or based on this formula: (Maximum Scale point – Minimum
Scale point)/4.

Along with the above ways/methods, we consider several factors

such as cost, time etc. while determining the sample size.
 Measurement and Scaling

Measurement means assigning numbers to characteristics of commercial objects.

Scaling involves creating a continuum upon which measured objects are located.

Scaling is an extension of measurement. Consider an attitude scale from 1 to 7. Each

respondent is assigned a number from 1 to 7, with 1 = Extremely Unfavorable, and 7 =
Extremely Favorable. Measurement is the actual assignment of a number from 1 to 7
to each respondent (e.g., 1, 2, 3, 4, 5, 6, 7). Scaling is the process of placing the
respondents on a continuum with respect to their attitude toward department stores
(i.e., (1-7)-point scale).

Scale is an instrument to measure and quantify consumer perception/

actual behavior

Types of Marketing Scale/Measure/Construct

Scale

Nominal Ordinal Interval Ratio

Categorical/Non-metric Non-categorical/Metric
Scale Scale

Types of scale to be used in data collection will decide appropriate data analytics
 Types of Marketing Scale/Measure/Construct

Scale Basic Common Marketing Permissible Statistics

Characteristics Examples Examples Descriptive Inferential
Nominal Numbers identify Social Security Brand nos., store Percentages, Chi-square,
& classify objects nos., numbering types mode binomial test
of football players
Ordinal Nos. indicate the Quality rankings, Preference Percentile, Rank-order
relative positions rankings of teams rankings, market median correlation,
of objects but not in a tournament position, social Friedman
the magnitude of class ANOVA
differences
between them
Interval Differences Temperature Attitudes, Range, mean, Product-
between objects opinions, index standard moment
Ratio Zero point is fixed, Length, weight Age, sales, Geometric Coefficient of
ratios of scale income, costs mean, harmonic variation
values can be mean
compared

Refer a common example with Flight in the next Slide… ☺

Types of Marketing Scale/Measure/Construct _ An example of Flight

Nominal Scale: The Flight number - Boeing 737-700BCF/BDSF

Ordinal Scale: Ranking of the Airline – 3rd

Interval Scale: Customer Satisfaction feedback on a 5-point scale: 1 = Strongly

dissatisfied ~ 5 = Strongly satisfied (difference between anchor points to be same)

Ratio Scale: Price of the tickets/foods such free foods (Fix zero)
 A Classification of Data Analysis Techniques
Data Analysis Techniques

Descriptive analysis Inferential analysis

Univariate Multivariate

A Classification of Univariate Techniques

Univariate Techniques

Metric Test (Data) Non-metric Test (Data)

One Sample Two Samples One Sample Two or More

Samples
* t test * Frequency
* Z test * Chi-Square
* K-S
* Runs
* Binomial
Independent Related
Independent Related
*Two sample t- * Paired
test t test * Chi-Square * Sign
* Mann-Whitney * Wilcoxon
* Median * McNemar
* K-S * Chi-Square
* K-W ANOVA
A Classification of Multivariate Techniques
Multivariate Techniques

Dependence Interdependence
Technique Technique

• ANOVA • Factor Analysis

• Discriminant analysis • Cluster Analysis
• Regression • Multidimensional Scaling
• Conjoint analysis

T-test and its family

Univariate Techniques

Metric Test (Data) Non-metric Test (Data)

Types No. of No. of Frequenc Examples

sampl variab y of data
One Sample Two e unit le collection
Samples
* t test
One sample 1 1 1 Test of average
customer satisfaction
rating is 3.5 (of 5
scale)
Two sample 2 1 1 Customer satisfaction
Independent Related (independent ) rating of females is
higher than males
*Two sample t- * Paired Paired/dependent/r 1 1 2 times Customer satisfaction
test t test epeated (e.g., before and after
before discount
and after
a
stimulus)-
widely
used
 Steps Involved in Hypothesis Testing
Formulate H0 and H1

Select Appropriate Test

Choose Level of Significance ()

Collect Data and Calculate Test Statistic

SPSS Manual
Look the Probability/Sig (p- Determine Critical Value
value) Associated with Test of Test Statistic TSCR.
Statistic.
Determine if TSCAL
Compare with Level of
Significance () with p- falls into (Non)
value Rejection Region
Reject or Do not Reject H0 or H1

Draw Marketing Research Conclusion

Ethnographic – Conceptualization

◼ Harvard Business Review defines ethnography as a “branch of anthropology

that involves trying to understand how people live their lives.”

◼ Ethnographic research is the study of human behaviour in its natural context

and involves observation of behaviour and setting along with depth
interviews.

◼ Sometimes, people can’t articulate what they’re looking for in products and
services. By understanding how people live, researchers discover otherwise
elusive trends that inform the company’s future strategies.
 Ethnographic – Live example of Tide Naturals

• Ethnographic research at P&G found that 80% of

consumers in India wash their clothes by hand.

• They had to choose between detergents that

were relatively gentle on the skin but not very
good at actually cleaning clothes, and more-
potent but harsher agents.

• With the problem clearly identified, in 2009 the

team came up with Tide Naturals, which cleaned
well without causing irritation.

Observation method – Conceptualization

Observation methods involve recording the behavioural patterns of

people, objects, and events in a systematic manner to obtain
information about the phenomenon of interest.

The observation method does not question or communicate with

the people being observed. Information may be recorded as the
events occur or from the records of the past events.
 Analytics to Test these Hypotheses?
Refer the following hypotheses and recommend the analytics to be used to test them:

H1: Average Sales of Big Bazaar in Bangalore is higher than the same in Delhi.

H2: Average Sales of Big Bazaar in Bangalore is higher than the same in Delhi and
Kolkata.

H3: Average Sales of Big Bazaar across the cities Bangalore, Delhi, Kolkata, Mumbai,
Chennai, and Hyderabad are different.

ANOVA – Nature of the Data (Main Assumption)

ANOVA (Dependence Analytic)

Independent Dependent
Variable (s) Variable

Categorical (Nominal or Metric (Interval or Ratio

Ordinal Scale) Scale)

Necessary condition of ANOVA.

 ANOVA – Types (ANOVA Family)

No. & Nature of IV No. & Nature of DV Types of test

ONE & categorical ONE metric ONE-way ANOVA
TWO & Categorical ONE metric TWO-way ANOVA
ONE categorical and ONE non-categorical ONE metric ANCOVA
(covariate)
ONE or MORE & categorical TWO or MORE metric MANOVA

ANOVA - Conceptualization

• Analysis of variance (ANOVA) is used as a test of means for two or more

populations (μ1, μ2, μ3……… μk).

H0: μ1 = μ2 = μ3 ... = μk
H1: At least one pair of means is not all equal (e.g., μ2 ≠ μ3)

Example: A marketing research consultant is interested to examine whether Heavy vs.

Medium vs. Light vs. Nonusers of cold drink differ in their Preference for the cold drink
Coca-Cola, measured in a 7-point Likert-type scale ( 1 = lowest preference ~ 7 = highest
preference). The H0 and H1 are as follows:

H0: The four groups are not different in preference.

H1: At least one pair of groups is different.