NOTES- MARKETING RESEARCH

Disclaimer: This material is provided for study assistance only. The author holds no
responsibility for accuracy, marks obtained, or outcomes. Readers are not compelled to
follow or rely on these notes.

Module 1 FOUNDATIONS OF MARKETING RESEARCH

Introduction to Marketing Research & Applications of Research Designs

Meaning & Role of Marketing Research

Meaning:
Marketing Research is the systematic and objective process of collecting, analyzing, and
interpreting data to help managers make informed marketing decisions.

Key Features:

• Systematic (follows a structured process)

• Objective (free from bias)

• Decision-oriented

• Continuous process

Role of Marketing Research:

• Identifies customer needs and preferences

• Reduces risk in decision-making

• Helps in product development

• Evaluates marketing performance

• Supports strategic planning

Example:
A company like a new FMCG brand conducts research before launching a shampoo to
understand customer preferences regarding price, fragrance, and packaging.

Role of Information in Marketing Decisions

Importance of Information:

• Acts as the foundation of decision-making

• Helps reduce uncertainty

• Provides insights into market trends

BUNNY <’3
 • Supports competitive advantage

Types of Information:

• Internal (sales, costs, inventory)

• External (market trends, competitors)

• Primary (collected for a specific purpose)

• Secondary (already available data)

Marketing Decisions Supported:

• Product decisions (features, design)

• Pricing decisions

• Promotion strategies

• Distribution channels

Example:
A smartphone company uses customer feedback data to decide whether to launch a budget or
premium model.

Marketing Research – Sequence of Steps

BUNNY <’3
Steps in Marketing Research Process:

1. Problem Definition

o Identify the problem clearly

o Example: Why are sales declining?

2. Development of Research Design

o Decide type of research (exploratory, descriptive, causal)

3. Data Collection

o Primary or secondary data

4. Data Analysis

o Use statistical tools to interpret data

5. Report Preparation & Presentation

o Present findings to management

6. Decision Making

o Take action based on insights

Example:
If a company notices declining sales, it conducts research to identify reasons such as poor
pricing or lack of promotion.

Meaning & Types of Research Designs

BUNNY <’3
Meaning:
Research design is the framework or blueprint for conducting marketing research.

Types of Research Designs

1. Exploratory Research

• Used when the problem is not clearly defined

• Flexible and unstructured

• Generates ideas and insights

Methods:

• Literature review

BUNNY <’3
 • Expert interviews

• Focus groups

Example:
Understanding why customers prefer a competitor brand.

2. Descriptive Research

• Describes characteristics of a market

• Structured and formal

• Answers who, what, when, where, how

Methods:

• Surveys

• Observation

Example:
Finding the percentage of customers who prefer online shopping.

3. Causal Research

• Identifies cause-and-effect relationships

• Uses experiments

Example:
Testing whether a discount increases sales.

Problem Formulation in Marketing Research

Meaning:
Problem formulation is the process of clearly defining the research problem.

Importance:

• Guides the entire research process

• Prevents wastage of time and resources

• Ensures relevant data collection

Types of Problems:

1. Management Decision Problem – What should management do?

2. Marketing Research Problem – What information is needed?

BUNNY <’3
Steps in Problem Formulation:

1. Identify management problem

2. Convert into research problem

3. Define objectives

4. Develop hypotheses

5. Identify variables

Example:

• Management Problem: Sales are declining

• Research Problem: Identify factors affecting customer purchase behaviour

Application of Qualitative Research in Marketing

Basic Introduction to Qualitative Research

Meaning:
Qualitative research focuses on understanding attitudes, opinions, and motivations.

Characteristics:

• Small sample size

• Non-structured

• In-depth insights

• Subjective analysis

Use Cases:

• New product ideas

• Brand perception

• Consumer behaviour

Qualitative Techniques

BUNNY <’3
1. Depth Interview

• One-to-one interview

• Unstructured

• Probes deeply into respondent’s thoughts

Example:
Interviewing a customer to understand why they prefer organic food.

2. Focus Group

• Group discussion (8–12 people)

• Moderated by a researcher

• Generates diverse opinions

Example:
Testing reactions to a new advertisement.

3. Projective Techniques

• Indirect questioning

• Helps uncover hidden motives

Types:

• Word association

• Sentence completion

• Picture interpretation

Example:
“What comes to your mind when you hear ‘luxury car’?”

Emerging Issues in Marketing Research

1. Marketing Research in the Internet Era

• Rapid data collection through digital platforms

• Real-time feedback

• Global reach

Examples:

• Online surveys

BUNNY <’3
 • Social media analysis

• Website analytics

2. Online Research

Meaning:
Collecting data through the internet.

Methods:

• Email surveys

• Online panels

• Mobile surveys

Advantages:

• Cost-effective

• Fast

• Large sample reach

Limitations:

• Sampling bias

• Low response quality

Data Warehousing and Data Mining

Data Warehousing:

• Central storage of large volumes of data

• Integrates data from multiple sources

Example:
Retail chains storing customer purchase data.

Data Mining:

• Extracting useful patterns from data

Techniques:

• Classification

• Clustering

• Association

BUNNY <’3
Example:
Identifying that customers who buy bread also buy butter.

Module 2 QUESTIONNAIRES & SCALING

Scale Construction and Purification Process

Meaning:
Scale construction is the process of developing a set of items to measure a concept (e.g.,
satisfaction, attitude).
Purification ensures that the scale is accurate, reliable, and valid.

Steps in Scale Construction

1. Define Construct

o What to measure (e.g., customer satisfaction)

BUNNY <’3
 2. Generate Items

o Create statements/questions

3. Choose Measurement Scale

o Likert, semantic differential, etc.

4. Design Questionnaire

5. Pilot Testing

o Test on small sample

6. Item Analysis

o Remove weak or confusing items

7. Reliability Testing

o Consistency check

8. Validity Testing

o Accuracy check

9. Final Scale Formation

Example:
Measuring satisfaction using statements like “I am satisfied with product quality.”

Designing Questionnaires for Market Research

Meaning:
A questionnaire is a structured set of questions used to collect data.

Objectives:

• Translate research objectives into questions

• Collect accurate and relevant data

• Ensure easy understanding

Principles of Good Questionnaire Design:

• Use simple and clear language

• Avoid leading or biased questions

• Keep it concise

• Logical sequence of questions

BUNNY <’3
 • Avoid ambiguity

Structure of Questionnaire:

1. Introduction

2. Screening Questions

3. Main Questions

4. Classification Questions (age, income)

5. Closing

Example:
Instead of asking “Do you like our excellent product?”, ask “How satisfied are you with the
product?”

Scales of Measurement Used in Marketing Research

BUNNY <’3
1. Nominal Scale

• Classification only

• No order

Example: Gender, Brand preference

2. Ordinal Scale

• Ranking or order

• No equal intervals

Example: Rank brands from best to worst

3. Interval Scale

• Equal intervals

• No true zero

Example: Temperature scale

4. Ratio Scale

• Equal intervals + true zero

Example: Income, age

Common Attitude Scales

• Likert Scale (Strongly agree → Strongly disagree)

• Semantic Differential Scale (Good – Bad)

• Stapel Scale (-5 to +5)

Structured and Unstructured Questionnaires

Structured Questionnaire:

• Fixed questions and options

• Easy to analyse

Example:
“How satisfied are you?”

• Very satisfied

• Satisfied

BUNNY <’3
 • Neutral

Unstructured Questionnaire:

• Open-ended questions

• Exploratory research

Example:
“What do you think about our product?”

Difference:

Basis Structured Unstructured

Format Fixed Flexible

Data Quantitative Qualitative

Analysis Easy Difficult

How to Choose a Scale and Question Type

Factors to Consider:

1. Research objective

2. Type of data required

3. Respondent ability

4. Level of accuracy needed

5. Time and cost

Guidelines:

Objective Suitable Scale

Measure attitude Likert scale

Compare brands Ranking scale

Measure perception Semantic differential

Collect facts Ratio scale

Example:
Customer satisfaction → Likert scale (1 to 5)

BUNNY <’3
Transforming Information Needs into a Questionnaire

Process:

1. Identify Information Needed

o Example: Customer satisfaction

2. Define Variables

o Quality, price, service

3. Operationalize Variables

o Convert into measurable items

4. Develop Questions

BUNNY <’3
 o “Rate product quality”

5. Select Scale

o Likert scale (1–5)

6. Design Questionnaire

Example:

• Need: Brand perception

• Question: “How would you rate the brand on quality?”

Application of Concepts (Scale Design Examples)

1. New Product Development & Concept Testing

Scale Example (Likert):

• “This product idea is appealing”

• “I would buy this product”

Scale: 1 (Strongly disagree) to 5 (Strongly agree)

2. Customer Satisfaction & Loyalty

Scale Example:

• “I am satisfied with the product”

• “I will repurchase this brand”

3. Brand Perception & Awareness

Semantic Differential Scale:

| Brand | Poor | 1 | 2 | 3 | 4 | 5 | Excellent |

4. Market Segmentation

Questions:

• Age, income (Ratio scale)

• Lifestyle (Likert scale)

5. Advertising Effectiveness & Recall

• “Have you seen this advertisement?” (Yes/No)

• “How memorable was the ad?” (1–5 scale)

BUNNY <’3
6. Pricing Research

• “At what price would you consider this product expensive?”

• “Is the product worth the price?”

7. Use of Conceptual and Theoretical Model

• Based on theories (e.g., attitude → intention → behaviour)

• Helps structure questionnaire logically

Example:
Attitude → Purchase intention → Actual buying

Reliability and Validity of a Questionnaire

Reliability

Meaning:
Consistency of measurement

Types:

• Test-retest reliability

• Internal consistency (Cronbach Alpha)

• Split-half reliability

BUNNY <’3
Example:
If the same person gives similar answers repeatedly → reliable scale

Validity

Meaning:
Accuracy of measurement

Types:

1. Content Validity – Covers all aspects

2. Construct Validity – Measures correct concept

3. Criterion Validity – Correlates with external criteria

Example:
If a satisfaction scale actually measures satisfaction → valid

Difference:

Basis Reliability Validity

Meaning Consistency Accuracy

Focus Stability Correctness

Relationship Reliability is necessary for validity

Module 4 Sampling and Fieldwork

Basics of Sampling & Types of Sampling

Sampling is the process of selecting a subset (sample) from a larger group (population) to
obtain information and draw conclusions about the entire population.

Terms in Sampling

• Population (Universe): Entire group of elements (e.g., all customers of a brand)

• Sample: Subset selected from the population

• Sampling Unit: Basic unit (individual, household, firm)

• Sampling Frame: List of all population elements

• Parameter: True value in population

• Statistic: Estimate derived from sample

BUNNY <’3
Why Sampling is Used

• Saves time and cost

• More practical and feasible

• Allows better control and accuracy in many cases

Sampling Process

1. Define the target population

2. Determine sampling frame

3. Select sampling technique

4. Decide sample size

5. Execute sampling process

Types of Sampling

A. Probability Sampling (Random Sampling)

Every element has a known and non-zero chance of selection.

1. Simple Random Sampling

BUNNY <’3
 • Each element has equal probability

• Example: Lottery method

2. Systematic Sampling

• Select every kth element

• Example: Every 10th customer

3. Stratified Sampling

• Population divided into homogeneous groups (strata)

• Samples taken from each group

4. Cluster Sampling

• Population divided into clusters

• Entire clusters selected randomly

B. Non-Probability Sampling

Probability of selection is unknown

1. Convenience Sampling

• Based on ease of access

• Example: Surveying people in a mall

2. Judgmental Sampling

• Researcher selects based on expertise

3. Quota Sampling

• Predefined quotas for categories

4. Snowball Sampling

• Existing respondents refer others

BUNNY <’3
Comparison of Sampling Types:

Basis Probability Sampling Non-Probability Sampling

Selection Random Non-random

Bias Less More

Generalization High Limited

Cost Higher Lower

Nature of Fieldwork

Fieldwork refers to the actual data collection process carried out by field investigators from
respondents.

Fieldwork / Data Collection Process

BUNNY <’3
 1. Selection of fieldworkers

2. Training of fieldworkers

3. Supervision

4. Validation

5. Evaluation

Selection of Cities / Centres:

• Depends on research objectives

• Should represent target population

• Factors considered:

o Population size

o Geographic spread

o Cost constraints

o Accessibility

Selection of Fieldworkers:

• Should have:

o Good communication skills

o Honesty and integrity

o Ability to understand questionnaire

o Basic education and language skills

BUNNY <’3
Training of Fieldworkers

BUNNY <’3
Purpose:

• Ensure uniformity and accuracy

Training Includes:

• Understanding objectives of research

• Detailed explanation of questionnaire

• Interview techniques

• Handling respondents

• Recording responses correctly

Supervision of Fieldworkers:

• Ensures quality control

• Activities include:

o Checking completed questionnaires

o Observing interviews

o Providing feedback

• Prevents fraud and errors

Validation of Fieldwork:

• Confirms whether data collected is accurate and genuine

• Methods:

BUNNY <’3
 o Re-contacting respondents

o Checking consistency

o Spot verification

Evaluation of Fieldworkers

• Based on:

o Quality of data collected

o Accuracy

o Completion rate

o Adherence to guidelines

• Helps in:

o Identifying best performers

o Improving future research

Module 4 DATA ANALYSIS – I

Introduction to Data Analysis

Data analysis is the process of transforming raw data into meaningful insights to support
decision-making. It involves organizing, summarizing, and interpreting collected data.

Objectives of Data Analysis

• Convert raw data into usable information

• Identify patterns, trends, and relationships

• Support managerial decision-making

• Test research hypotheses

Stages in Data Analysis

BUNNY <’3
 1. Data Cleaning

2. Data Validation

3. Data Coding (if required)

4. Data Entry

5. Data Analysis & Interpretation

Data Cleaning

Data cleaning is the process of detecting and correcting errors or inconsistencies in data.

Common Issues in Data

• Missing values

• Incomplete responses

• Outliers

• Incorrect entries

• Duplicate data

Methods of Data Cleaning

• Editing: Checking questionnaires for completeness and accuracy

• Handling Missing Data:

o Deletion method

o Imputation (mean/median substitution)

• Outlier Detection: Identifying abnormal values

• Consistency Check: Ensuring logical answers

Importance

• Improves accuracy and reliability

• Ensures valid results

• Prevents misleading conclusions

Data Validation

Data validation ensures that the collected data is authentic, accurate, and logically consistent.

Types of Validation

BUNNY <’3
 • Field Validation: Checking data during collection

• Post-Collection Validation: Checking after data entry

• Re-contact Method: Verifying responses with respondents

Validation Checks

• Range checks (e.g., age cannot be negative)

• Logical checks (e.g., income vs occupation)

• Consistency across responses

Data Entry in Analysis Tools

Purpose

To convert cleaned data into a structured format suitable for analysis using software tools.

1. Data Entry in Microsoft Excel

Features

• Simple and widely used

• Suitable for small to medium datasets

Process

• Rows represent respondents

• Columns represent variables/questions

BUNNY <’3
 • Use numerical coding (e.g., Male = 1, Female = 2)

• Apply filters and basic formulas

Advantages:

• Easy to use

• Cost-effective

• Supports basic analysis and charts

Data Entry in IBM SPSS Statistics

Features

• Advanced statistical analysis tool

• Suitable for large and complex datasets

Structure in SPSS

• Variable View:

o Define variable name, type, labels, values

BUNNY <’3
 • Data View:

o Enter actual data

Process:

1. Define variables in Variable View

2. Assign codes and labels

3. Enter data in Data View

4. Save dataset for analysis

Advantages:

• Handles complex analysis

• Reduces manual errors

• Provides statistical outputs directly

Comparison: Excel vs SPSS

Basis Excel SPSS

Ease of Use Very Easy Moderate

Data Size Small–Medium Large

Analysis Basic Advanced

Coding Support Limited Strong

DATA ANALYSIS – Interpretation with Detailed Case & Excel-Type Tables

Case Study (Common for All Tests)

A company collected data from 120 customers to analyze:

• Satisfaction score (1–10)

• Gender (Male/Female)

• City (Pune, Mumbai, Delhi)

• Income (₹)

• Advertisement exposure (Low/Medium/High)

• Before–After campaign satisfaction

BUNNY <’3
Objective: Test differences, relationships, and predictions

UNIVARIATE ANALYSIS

Excel-Type Output Table (Descriptive Statistics)

Statistic Satisfaction

Mean 7.20

Median 7.00

Mode 8

Std. Deviation 1.50

Min 3

Max 10

Interpretation:

The average satisfaction score is 7.20, indicating that customers are generally satisfied.
The standard deviation of 1.50 shows moderate variation in responses.

BIVARIATE ANALYSIS (Cross Tabulation)

Excel-Type Cross Tab Table

BUNNY <’3
 Gender High Satisfaction Low Satisfaction Total

Male 30 20 50

Female 50 20 70

Total 80 40 120

Interpretation:

Females have a higher number of satisfied customers (50) compared to males (30), indicating
a possible relationship between gender and satisfaction.

CHI-SQUARE TEST (TEST OF ASSOCIATION)

Excel Output Table

BUNNY <’3
 Statistic Value

Chi-Square 4.80

df 1

p-value 0.028

Interpretation (Step Format)

• Chi-square test applied to check association between gender and satisfaction

• p-value = 0.028 < 0.05

• Reject H₀

• There is a significant relationship between gender and satisfaction

t-TESTS

A. One Sample t-Test

Excel Output Table

Statistic Value

Mean 7.20

BUNNY <’3
 Test Value 7

t-value 2.10

p-value 0.037

Interpretation:

• p-value < 0.05

• Mean satisfaction is significantly different from 7

B. Independent Sample t-Test

Excel Output Table

Group Mean

Male 6.50

Female 7.80

Statistic Value

t-value 3.20

p-value 0.002

Interpretation

• p-value < 0.05

• Significant difference exists

BUNNY <’3
 • Females are more satisfied than males

Paired Sample t-Test

Excel Output Table

Condition Mean

Before 6.80

After 7.90

Statistic Value

t-value 4.50

p-value 0.000

Interpretation

• p-value < 0.05

• Advertising campaign significantly increased satisfaction

ANOVA

Excel Output Table

BUNNY <’3
 Source SS df MS F p-value

Between Groups 30 2 15 5.60 0.005

Within Groups 200 117 1.71

Total 230 119

Interpretation

• p-value = 0.005 < 0.05

• Reject H₀

• Satisfaction differs significantly across cities

MANOVA (Concept + Output Style)

Statistic Value

Wilks’ Lambda 0.85

p-value 0.012

Interpretation

• p-value < 0.05

BUNNY <’3
 • Cities differ significantly on combined variables (satisfaction & loyalty)

CORRELATION

Excel Output Table

BUNNY <’3
 Variables Satisfaction Income

Satisfaction 1 0.70

Income 0.70 1

Interpretation

• r = 0.70

• Strong positive relationship

• Higher income → higher satisfaction

REGRESSION

Excel Output Table (Simple Regression)

Model Summary

R R²

0.70 0.49

Coefficients Table

Variable Coefficient p-value

Constant 2.5 0.01

Income 0.004 0.000

BUNNY <’3
Interpretation

• R² = 0.49 → 49% variation explained

• Income significantly affects satisfaction

• Positive coefficient → higher income increases satisfaction

Multiple Regression:

Variable Coefficient p-value

Income 0.003 0.001

Advertising 0.80 0.000

Price -0.50 0.02

Interpretation

• Advertising has strongest positive impact

• Price negatively affects satisfaction

DISCRIMINANT ANALYSIS

Output Table

Classification Table:

Actual / Predicted High Low

High 45 5

Low 10 40

Accuracy = 85%

Interpretation

• Model correctly classifies 85% cases

BUNNY <’3
 • Variables used are effective in distinguishing groups

STANDARD INTERPRETATION FORMAT:

1. Test Applied

2. Result (p-value / statistic)

3. Decision (Reject/Accept H₀)

4. Conclusion (Managerial Meaning)

Example

ANOVA was applied to test differences in satisfaction across cities. The p-value (0.005) is
less than 0.05, so the null hypothesis is rejected. Therefore, there is a significant difference in
satisfaction among cities.

Module 5 DATA ANALYSIS-II

FACTOR ANALYSIS

Factor Analysis is a data reduction technique used to identify underlying factors (latent
variables) that explain correlations among multiple observed variables.

Example:
10 product attributes → reduced to 2–3 key factors like Quality, Price Sensitivity

Data Requirements:

• Sample size: Minimum 5–10 respondents per variable

• Variables should be correlated

• Use metric data (interval/ratio preferred)

Pre-Analysis Tests (Excel/SPSS Output Style)

KMO and Bartlett’s Test

BUNNY <’3
 Test Value

KMO 0.78

Bartlett’s Sig. 0.000

Interpretation

• KMO > 0.5 → Sample adequate

• Sig. < 0.05 → Variables are correlated

• Factor analysis is appropriate

EFA vs CFA

Basis EFA CFA

Purpose Explore structure Confirm structure

Prior knowledge Not required Required

Usage Initial research Theory testing

BUNNY <’3
Factor Extraction (Total Variance Explained)

Factor Eigenvalue % Variance

1 3.5 35%

2 2.2 22%

Total 57%

Interpretation

• Only factors with Eigenvalue > 1 are selected

• 2 factors explain 57% variance

Factor Matrix (Rotated Component Matrix)

BUNNY <’3
 Variables Factor 1 Factor 2

Durability 0.82 0.10

Reliability 0.79 0.15

Price 0.12 0.75

Discount 0.20 0.72

Interpreting & Naming Factors

Interpretation

• Factor loadings > 0.5 are significant

• Group variables with high loadings

Naming

• Factor 1 → Product Quality

• Factor 2 → Price Sensitivity

CLUSTER ANALYSIS

Cluster analysis groups respondents into homogeneous segments based on similarity.

Procedure

1. Select variables

2. Choose clustering method

3. Measure similarity (distance)

4. Form clusters

5. Interpret clusters

Types of Clustering

A. Hierarchical Clustering

• Builds clusters step-by-step

• Output: Dendrogram

B. Non-Hierarchical (K-Means)

• Pre-defined number of clusters

BUNNY <’3
 • More efficient for large data

Dendrogram (Hierarchical Output):

Interpretation

• Cutting tree at a level gives number of clusters

• Example: 3 distinct customer segments

K-Means Output Table

BUNNY <’3
Final Cluster Centers

Variable Cluster 1 Cluster 2

Income High Low

Price Sensitivity Low High

Interpretation

• Cluster 1 → Premium customers

• Cluster 2 → Price-sensitive customers

CONJOINT ANALYSIS

Conjoint analysis determines how consumers value different product attributes.

Example:
Mobile phone → Price, Brand, Battery, Camera

Procedure

BUNNY <’3
 1. Identify attributes & levels

2. Create product profiles

3. Collect preference data

4. Estimate utility values

5. Determine importance

Type of Data Required

• Preference ratings or rankings

• Trade-off data

Output Table (Utility Scores)

BUNNY <’3
 Attribute Level Utility

Price Low 2.5

Price High -1.2

Brand A 1.8

Brand B 0.5

Interpretation

• Higher utility = higher preference

• Price has strong influence

• Consumers prefer low price & Brand A

REPORT WRITING

Standard Report Format:

1. Title Page

2. Executive Summary

BUNNY <’3
 3. Introduction

4. Research Objectives

5. Methodology

6. Data Analysis

7. Findings

8. Conclusion

9. Recommendations

10. Appendix

Guidelines for Report Writing

• Clear and simple language

• Logical flow

• Avoid technical jargon

• Use charts and tables

Guidelines for Tables

Example Table Format

Factor Loading

Quality 0.82

Price 0.75

Rules

• Proper title

• Units clearly mentioned

• Avoid overcrowding

Guidelines for Graphs:

• Use appropriate graph type

• Label axes clearly

• Maintain simplicity

• Highlight key insights

BUNNY <’3
Presentation Tips

• Use bullet points

• Highlight key findings

• Focus on decision-making insights

BUNNY <’3