Business Research Methods

Module I: Nature and Scope of Research Methodology

1. Types of Research

 Exploratory Research – Used to explore a problem with no clear solution (e.g., interviews, focus
groups). Helps generate new ideas for further study.
 Conclusive Research – Used for decision-making, further divided into:
o Descriptive Research – Describes characteristics, trends, or patterns (e.g., surveys, case
studies). It does not establish cause and effect.
o Causal Research – Establishes cause-and-effect relationships by manipulating variables
(e.g., A/B testing, experiments).

2. Research Process and Steps

Step Explanation Example

Identify the Research Define the issue or gap in knowledge. Lack of customer engagement in
Problem an app.
Review the Literature Study previous research for insights Reading published papers on
and gaps. customer behavior.
Formulate Make predictions or set research Hypothesis: “Discounts increase
Hypothesis/Objectives goals. sales.”
Choose Research Design Select method (Exploratory, Conducting surveys or
Descriptive, Experimental). experiments.
Collect Data Use surveys, interviews, Sending online forms or
observations, etc. conducting focus groups.
Analyse Data Process and interpret data using Using Excel/SPSS for data
statistical tools. analysis.
Interpret and Conclude Make sense of findings and test “Discounts increase sales by
hypothesis. 20%.”
Report and Present Share research results with Writing reports or making
stakeholders. presentations.

3. Approaches to Research

Approach Meaning Example

Deductive Starts with a theory and tests it. Hypothesis testing in experiments.
Inductive Observations lead to a new theory. Case studies, qualitative research.
Qualitative Non-numerical data, subjective analysis. Interviews, focus groups.
Quantitativ Numerical data, objective analysis. Surveys, experiments.
e

4. Research Design

 Exploratory Design – Open-ended research used to explore ideas before forming hypotheses (e.g.,
focus groups, interviews, secondary data).
 Descriptive Design – Focuses on describing characteristics of variables without manipulating them
(e.g., surveys, case studies).
 Experimental Design – Involves manipulating variables to test cause-and-effect relationships (e.g.,
clinical trials, A/B testing).
 Module II: Research Methods and Data Collection Techniques

1. Research Modelling

A research model is a framework that guides the study.

 Stages of Research Modelling:

1. Define the problem
2. Develop the model (theoretical or empirical)
3. Select data collection methods
4. Analyse the data
5. Interpret findings
 Types of Research Models:

o Conceptual Model – Based on theories, no direct data testing.

o Empirical Model – Uses real-world data to validate hypotheses.

2. Data Collection Methods

Method Description Example

Survey Asking structured questions to a Online customer feedback forms.
sample.
Observation Watching and recording behavior. CCTV analysis in retail stores.
Questionnaire Set of written questions for Employee satisfaction survey.
respondents.
Focus Group Small group discussions to gather 6-10 people discussing new product
Discussions opinions. ideas.

3. Questionnaire Design

Steps: Define objectives → Select question type → Sequence logically → Pilot test → Finalize.
Types of Questions: Open-ended, close-ended, Likert scale, multiple-choice.

4. Attitude Measurement & Scaling Techniques

Scale Meaning Order Equal True Examples

Gaps Zero
Nomina Used for labeling or ❌ No ❌ No ❌ No Gender, Blood group, Marital
l categorizing data Status
Ordinal Shows rank/order, but no ✅ Yes ❌ No ❌ No Satisfaction level
exact differences (High/Med/Low), Ranks
Interval Shows order + equal intervals, ✅ Yes ✅ Yes ❌ No Temperature in °C, IQ scores
but no true zero
Ratio Has all features: order, equal ✅ Yes ✅ Yes ✅ Yes Height, Weight, Age, Income
gaps & true zero

 Nominal = Name only

 Ordinal = Order matters
  Interval = Interval is equal
 Ratio = Real Zero exists

Scale Description Example

Nomina Categories without order, used for Gender (Male/Female), Blood Group
l classification. (A/B/O).
Ordinal Ordered categories, but differences between Customer satisfaction levels (High, Medium,
ranks are not equal. Low).
Interval Ordered, equal intervals, no true zero. Can Temperature in Celsius (20°C to 30°C = 10°C
perform addition and subtraction. increase).
Ratio Ordered, equal intervals, has a true zero. Height, weight, age, income (Zero means
Allows all mathematical operations. absence of value).

5. Sampling Plan

Sampling Type Description Types

Probability Every unit in the 1. Simple Random Sampling – Every individual has an equal
Sampling population has a chance.
known chance of 2. Stratified Sampling – Population divided into groups (strata),
selection. then samples taken from each.
More accurate & 3. Cluster Sampling – Population divided into clusters, then a
unbiased. few clusters are selected randomly.
4. Systematic Sampling – Every nth element from a list is
chosen.
Non- Selection is based 1. Convenience Sampling – Easily accessible participants are
Probability on convenience or selected.
Sampling judgment, not 2. Judgmental Sampling – Researcher selects participants
random. based on expertise.
More biased but 3. Snowball Sampling – Existing participants recruit more
useful for respondents.
exploratory 4. Quota Sampling – Population divided into quotas, and a
research. specific number of responses are taken.

6. Errors in Research

 Sampling Errors – Differences between sample results and actual population data. Reduced by
increasing sample size.
 Non-Sampling Errors – Errors due to faulty questionnaire, non-response, data entry mistakes,
interviewer bias.

7. Editing, Tabulating, and Validating Data

 Editing – Checking for mistakes in data collection and correcting errors.

 Tabulating – Organizing data in tables for easy analysis and comparison.
 Validating – Ensuring accuracy and consistency before analysis by checking completeness,
correctness, and reliability of responses.

Module III: Data Analysis Techniques

1. Statistics – Meaning
  Definition: Statistics is the science of collecting, organizing, analyzing, and interpreting numerical
data.
 Use in Research: It helps researchers draw conclusions and make decisions based on data.
 Example: A company uses sales data from last year to plan for next year’s inventory.

2. Types of Statistics

There are two main types:

Type Meaning Example

Descriptive Describes and summarizes data in a simple way. Average marks of students in a
Statistics class.
Inferential Helps to make predictions or generalizations about Predicting election results using
Statistics a population from a sample. sample surveys.

3. Descriptive Statistics – Meaning

 Definition: It is the process of summarizing or describing a set of data.

 Tools Used:
o Measures of Central Tendency (Mean, Median, Mode)
o Measures of Dispersion (Range, Variance, Standard Deviation)
o Tables, Graphs, Charts (like pie charts, bar graphs)

Example:
A survey shows that the average salary of MBA graduates in Delhi is ₹6 LPA.

4. Inferential Statistics – Meaning

 Definition: It allows researchers to make predictions or inferences about a population based on a

sample.
 Use: Helps in decision-making under uncertainty.
 Tools Used:
o Hypothesis Testing
o Confidence Intervals
o Regression Analysis
o ANOVA, Chi-square test

Example:
Surveying 500 people and using that sample to predict the opinions of 10,000 people.

Key Differences: Descriptive vs Inferential Statistics

Basis Descriptive Statistics Inferential Statistics
Purpose To describe data To draw conclusions from data
Scope Works on entire data Works on sample to make generalizations about
(population/sample) the population
Techniques Mean, Median, Mode, Graphs Hypothesis testing, Regression, Confidence
Used intervals
Example Showing average height of students Predicting average height of students in a city
in a class

Review of Hypothesis Testing Procedures

What is Hypothesis Testing?

 It is a statistical method used to make decisions or inferences about population parameters based on
sample data.
 It helps us test assumptions (called hypotheses) about a population.

Key Terms to Know

 Hypothesis: Assumption or claim about a population.

 Null Hypothesis (H₀): The default or original assumption (e.g., there is no effect/difference).
 Alternative Hypothesis (H₁ or Hα): Opposite of the null; the claim we want to test.
 Significance Level (α): The probability of rejecting H₀ when it is actually true (commonly 0.05 or
5%).
 p-value: Probability of getting the observed result assuming H₀ is true.

Steps in Hypothesis Testing

Ste Description
p
1️ State the Hypotheses: Define H₀ and H₁ clearly.
2️ Select Significance Level (α): Common choices are 0.05 or 0.01.
3️ Choose the Right Test: Z-test, t-test, chi-square, ANOVA, etc.
4️ Collect and Analyze Data: Calculate test statistic based on sample data.
5️ Compare with Critical Value OR use p-value: Decide whether to reject
H₀.
6️ Make a Decision: Conclude whether or not to reject the null hypothesis.
Example

Case: A company claims that the average delivery time is 3 days. A customer believes it is more than 3
days.

 H₀: μ = 3
 H₁: μ > 3
 Test used: One-sample t-test
 If p-value < 0.05 → Reject H₀ → The average delivery time is more than 3 days

Decision Criteria Table

p-value Decision Conclusion
p ≤ α (e.g., Reject H₀ There is enough evidence against
0.05) H₀
p>α Do not reject Not enough evidence to reject H₀
H₀

Tip to Remember

Reject H₀ if p is low → “p is low, let it go” (reject the null hypothesis)

Parametric and Non-Parametric Tests

What are Parametric Tests?
  These are statistical tests based on assumptions about the population parameters (like mean,
standard deviation).
 They require the data to be normally distributed.
 Used for quantitative (numerical) data.

Common Parametric Tests

Test Name Purpose Used When
Z-Test Tests the mean of a large sample (n > 30) Population variance is known
T-Test Tests the mean of a small sample (n ≤ 30) Population variance is unknown
F-Test Compares two or more variances Used before ANOVA or to test variance
equality

Example:
Comparing average marks of two MBA sections → use t-test if sample size is small.

What are Non-Parametric Tests?

 These tests do not assume a normal distribution or specific parameters.

 Used when data is ordinal, categorical, or ranked.
 Helpful when sample sizes are small or when data doesn’t meet parametric test conditions.

Common Non-Parametric Tests

Test Name Purpose Used When

Chi-square Test Tests the relationship between two For independence or
categorical variables goodness-of-fit
One-way ANOVA (non- Compares more than 2 groups on one Like Kruskal-Wallis Test
parametric) factor
Two-way ANOVA (non- Compares more than 2 groups on two Like Friedman Test
parametric) factors

Example:
Checking if gender (male/female) and preference for product type are related → use Chi-square test

Key Differences: Parametric vs Non-Parametric Tests

Basis Parametric Tests Non-Parametric Tests
Assumptions Based on population parameters and No strict assumptions about data distribution
normality
Type of Numerical (interval/ratio scale) Ordinal, nominal or ranked data
Data
Accuracy More powerful if assumptions are met Less powerful but more flexible
Examples Z-test, T-test, F-test Chi-square, Kruskal-Wallis, Friedman Test

Tip to Remember

 Parametric = Precise, Powerful, Population-based

 Non-Parametric = No Assumptions, Nominal/Ordinal data
 Module IV –
Field Project and Report

1. Pre-Writing Considerations
Before writing a research report, consider the following:
 ✅ Know your audience: Understand who will read the report (teachers, managers, clients).
 ✅ Define objective: What is the purpose? (To inform, analyze, recommend?)
 ✅ Collect reliable data: Ensure data is accurate, relevant, and complete.
 ✅ Choose a format/style: Formal tone, standard report format, consistent structure.
 ✅ Organize content: Create an outline of sections before you begin writing.

2. Research Report Components

Section Details
Title Page Report title, name of researcher, institution, date.
Table of Contents Lists all sections and page numbers.
Executive Summary / Abstract A short summary of the report (main points, findings,
recommendations).
Introduction Background, problem statement, objectives, and scope of study.
Literature Review Summary of previous research and theories related to the topic.
Research Methodology Details on methods used – research design, sampling, tools, data
collection techniques.
Data Analysis & Interpretation Tables, graphs, charts + explanation of findings.
Findings & Discussions Key results and what they mean in simple terms.
Conclusion Summary of insights gained from research.
Recommendations Actionable suggestions based on findings.
References/Bibliography List of sources and citations.
Appendices Extra materials like questionnaires, raw data, charts etc.

3. Common Problems Encountered When Preparing the Research Report

Problem Explanation
Lack of clarity Confused ideas, poor flow of content.
Poor organization Mixing sections, no logical structure.
Too much technical jargon Makes it hard for readers to understand.
Incomplete data Leads to weak analysis or wrong conclusions.
Grammatical & formatting Reduces report quality and readability.
errors
Plagiarism Copy-pasting content without credit is unethical.

4. Presenting the Research Report

Short pointers for effective presentation:

 🎯 Know your report well – Understand your key findings and data.
 Use visuals – Include charts, graphs, and tables to make it engaging.
 📢 Speak clearly and confidently – Practice before presenting.
 📝 Summarize key points – Don’t read the entire report word-by-word.
 💬 Handle questions smartly – Be honest, stay calm, and refer to data if unsure.
 👔 Maintain professional presentation – Use a clean format and formal tone.