Research Methods

Comprehensive Exam Guide — MBA Financial Management / Business Research

Compiled from NUML Quetta Course Lecture Slides

Covers: Introduction to Research · Literature Review · Research Design · Sampling & Sample Size ·
Data Collection & Measurement · Modern Data Collection Tools · Research Ethics

Prepared for Muhammad · July 2026

Table of Contents
Unit 1 — Introduction to Research Methodology
Definition & importance · Types of research · The 7-step research process · Validity vs reliability · Ethics basics · Case studies

Unit 2 — Literature Review & Research Problem Formulation

Sources of literature · Searching & Boolean operators · Reviewing literature · 6 types of literature review · Research gaps & problems · Objectives &
hypotheses

Unit 3 — Research Design

Exploratory, Descriptive, Experimental, Causal designs · Design choice by MBA specialization

Unit 4 — Sampling Techniques & Sample Size Determination

Population vs sample · Probability & non-probability sampling · Sample size methods & formulas · Alpha/Beta errors · Sampling challenges

Unit 5 — Data Collection Methods & Measurement/Scaling Techniques

Primary vs secondary data · Qualitative vs quantitative · Surveys, interviews, focus groups · Observation & case study · Measurement scales
(NOIR) · Reliability & validity · Questionnaire design

Unit 6 — Modern Data Collection Tools & Techniques

Online surveys & mobile tools · Social media analytics · Web scraping & big data · IoT & sensors · Traditional vs modern · Data-protection laws
(GDPR, HIPAA, PDPB)

Unit 7 — Ethical Considerations in Modern Data Collection

Six core ethical principles · Pakistan's legal framework · Ethical writing in a thesis · Ethical research checklist

Appendix
Master glossary · Formula sheet · Thesis-chapter roadmap · High-yield exam question bank
 Unit 1 — Introduction to Research Methodology
Foundational concepts every exam question builds on

1.1 What is Research?

Definition: Research is a systematic and organized effort to investigate a specific problem that requires a solution. It involves gathering
information, analyzing it, and drawing conclusions based on the findings.

Why Research Matters in Business

Informed Decision-Making — provides data-grounded (not assumption-based) decisions.
Competitive Advantage — reveals market trends, customer behavior, competitive dynamics.
Problem Solving — identifies problems and explores solutions.
Innovation & Growth — supports new products, services, and strategies.

1.2 Types of Research

A. Based on Objective

Type Purpose Example

Exploratory Problem not well defined; explores new insights, generates hypotheses Investigating customer behavior in a new market

Descriptive Describes characteristics of a phenomenon or relationship between Surveying customer satisfaction or employee
variables engagement

Analytical Explains causes and effects of phenomena/relationships Analyzing the effect of advertising on sales
(Explanatory)

Applied Practical research solving specific real-world problems Improving operational efficiency in manufacturing

B. Based on Methodology

Type Description

Quantitative Numerical data, statistical analysis; tools: surveys, experiments, data analytics

Qualitative Non-numerical data (interviews, observations, case studies); understands deeper meanings, behaviors, experiences

C. Based on Data Collection Method

Type Description

Primary Original data collected directly (surveys, interviews, experiments); directly relevant to the problem

Secondary Existing data (reports, articles, papers); cost-effective and time-saving but may not fully fit the research needs

Exam-relevant addition — Research Paradigms (not in slides, commonly tested)

Quantitative and qualitative research are grounded in different research philosophies:
Positivism — reality is objective and measurable; associated with quantitative, deductive research.
Interpretivism / Constructivism — reality is socially constructed and subjective; associated with qualitative, inductive
research.
Pragmatism — uses whichever methods best answer the question; underpins mixed-methods research (combining qualitative
+ quantitative).

1.3 The Research Process (7 Steps)

# Step Description

1 Problem Identification Recognize and define the problem/opportunity. E.g., "How does customer satisfaction affect brand loyalty?"

2 Literature Review Review existing literature to identify what is known and gaps. Sources: journals, books, industry reports, JSTOR, Google Scholar

3 Hypothesis Formulation A testable statement about the relationship between variables. E.g., "Higher satisfaction leads to greater loyalty."

4 Research Design Plan sampling, methods (surveys/interviews), and data collection approach

5 Data Collection Gather data using planned methods; ensure validity & reliability

6 Data Analysis Analyze data to test hypotheses/answer questions. Techniques: descriptive statistics, regression, t-tests

7 Conclusion & Reporting Draw conclusions; prepare and communicate a report to stakeholders
 1.4 Data Validity vs Data Reliability

Validity
Whether a measure actually captures what it is intended to measure. Example: A satisfaction survey asking about screen size and
camera is valid for measuring smartphone satisfaction; asking unrelated questions weakens validity.

Reliability
Whether a measure produces consistent results across repeated attempts. Example: If students retake a standardized math test
weeks apart and scores stay consistent, the test is reliable.

Memory tip: Validity = accuracy ("measuring the right thing"). Reliability = consistency ("measuring it the same way every time"). A test can be reliable without
being valid, but not valid without being reliable.

1.5 Ethical Considerations in Research (Overview)

Informed Consent — participants understand the purpose and voluntarily agree to participate.
Confidentiality — protect participant privacy; keep data confidential.
Avoidance of Bias — researchers strive for objectivity, avoid manipulating data toward a desired outcome.
Integrity — report findings honestly, even if results contradict the hypothesis.
Plagiarism — never use others' work/ideas without proper citation.

(A full, exam-focused treatment of ethics in modern digital research appears in Unit 7.)

1.6 Real-World Application (Case Studies)

Starbucks
Used customer feedback and market research (surveys, purchase-pattern analysis) to guide product launches, e.g., seasonal drinks,
and pricing strategy.

Coca-Cola Pakistan
Used surveys, focus groups, and data analytics to learn that Pakistani consumers preferred sweeter flavors and smaller, more
affordable packaging (especially in rural areas), and identified seasonal demand spikes (Ramadan, summer) — leading to targeted
campaigns and regional flavor innovations.

✅ Likely Exam Questions — Unit 1

1. Define research and explain its importance in business decision-making.
2. Differentiate exploratory, descriptive, analytical, and applied research with examples.
3. Distinguish quantitative and qualitative research.
4. List and explain the seven steps of the research process.
5. Differentiate between validity and reliability, with examples.
6. List the key ethical principles a researcher must follow.
7. Using the Starbucks/Coca-Cola case, explain how market research informs business decisions.
 Unit 2 — Literature Review & Research Problem Formulation
How to find, evaluate, and use existing research to build your own study

2.1 What is a Literature Review?

A critical and comprehensive analysis of existing research on a topic — systematically identifying, evaluating, and synthesizing prior work
to understand what is known and what gaps exist.

Objectives
Understand the current state of knowledge in a field
Identify key theories, concepts, and researchers
Determine gaps in existing research
Support formulation of research questions and hypotheses

2.2 Sources of Literature

Source
Definition Examples
Type

Primary First-hand, original research/data, not filtered or interpreted by Original research articles, case studies (e.g., HBR case on Netflix),
others experiments, theses/dissertations, patents

Secondary Interpretation, synthesis, or analysis of primary sources Literature reviews, systematic reviews, meta-analyses,
textbooks/encyclopedias

Tertiary Tools that index, organize, or compile primary/secondary sources Indexes, abstracts, databases (Scopus, Google Scholar, ProQuest, JSTOR)
— used to locate information

Summary Table

Source Type Use in Research Example Resource

Primary Gather original data Research papers, case studies

Secondary Analyze/summarize primary sources Literature & systematic reviews

Tertiary Help locate sources Google Scholar, Scopus

About Scopus: A large, comprehensive academic database covering science, technology, medicine, and social sciences. It indexes peer-reviewed journals,
tracks citations, and shows author profiles — useful for finding reliable sources and tracking research trends.

2.3 Searching for Literature

Key Steps
1. Define the research question — clarifies scope and guides keyword selection
2. Identify relevant keywords and synonyms
3. Use Boolean operators to build a search string

Operator Function Example

AND Narrows results marketing AND India

OR Broadens results marketing OR branding

NOT Excludes terms marketing NOT digital

Worked example: Research question — "How does social media marketing influence consumer buying behavior in Pakistan's retail
market?" →
Search string: ("Social media marketing" OR "digital marketing") AND ("consumer behavior" OR "buying behavior") AND ("retail industry"
OR "e-commerce") AND Pakistan

Tools: Google Scholar, JSTOR, Scopus, Web of Science, PubMed, ProQuest, SSRN, ResearchGate

2.4 Reviewing the Literature

1. Skim abstracts and conclusions to get main ideas quickly
2. Evaluate quality and credibility — is it peer-reviewed? Is the methodology solid?
3. Compare findings across studies — where do studies agree/differ?
4. Identify patterns, contradictions, and gaps
5. Organize findings — by theme (methods, outcomes) or chronologically
 2.5 Types of Literature Reviews

Type Definition Key Features Example

1. Traditional / General summary and critical discussion without a Subjective, broad; good for Review of leadership theories over 50 years
Narrative formal search/selection method theoretical grounding

2. Systematic Structured, transparent, reproducible method for Follows a protocol; minimizes bias; Reviewing 30 studies on digital marketing's
(SLR) identifying/evaluating all relevant studies inclusion/exclusion criteria effect on consumer behavior

3. Integrative Combines theoretical and empirical literature for a Synthesizes diverse viewpoints; Combining psychological + economic theories
comprehensive synthesis suited to theory development of decision-making

4. Theoretical Focuses on theories, models, frameworks underlying Builds/critiques theory; forms new Reviewing psychological theories to build a
a topic hypotheses new consumer-decision framework

5. Takes a clear stance, evaluating literature to Builds a strong argument; critical of Digital vs traditional marketing effectiveness
Argumentative support/challenge a position strengths/weaknesses

6. Critical Summarizes AND critically evaluates Deep methodology analysis; Critiquing CSR research methodologies
strengths/weaknesses of existing research highlights gaps; questions validity

Most common for MBA research

Systematic Literature Review (rigorous, minimizes bias — common in marketing, finance, business models); Integrative Review
(synthesizes theory + practice — common in leadership/organizational behavior); Critical Review (evaluates quality/gaps — common
in business ethics, governance, sustainability).

2.6 Identifying Research Gaps & Formulating the Research Problem

Research Gap: An area with a lack of information, conflicting evidence, or unanswered questions in existing literature.

How to identify gaps: look for inconsistencies/contradictions, understudied topics, limitations of previous studies, outdated research.

Research Problem: A specific issue, difficulty, contradiction, or gap in knowledge that the research intends to address.

Characteristics: clear and concise, researchable, relevant to the field, supported by literature.

Example: "Despite advancements in mobile banking, rural populations in Pakistan continue to rely on cash. Why?"

2.7 Research Objectives & Hypotheses

Research Objectives

Statements describing what the researcher aims to achieve.

General Objective — broad purpose

Specific Objectives — detailed, focused goals

SMART Criteria: Specific · Measurable · Achievable · Relevant · Time-bound

General: "To study the impact of social media marketing."

Specific: "To examine how Instagram ads influence purchase decisions of Gen Z in urban Pakistan within the past 2 years."

Hypotheses

A hypothesis is a tentative statement predicting a relationship between variables.

Null Hypothesis (H₀): no relationship exists

Alternative Hypothesis (H₁): a relationship exists

H₀: Social media ads have no effect on consumer purchase decisions.

H₁: Social media ads significantly influence consumer purchase decisions.

Exam-relevant addition — Variables (not explicit in slides, but essential for writing hypotheses)
Independent variable (IV): the presumed cause; manipulated or measured input (e.g., "advertising spend").
Dependent variable (DV): the presumed effect/outcome being measured (e.g., "sales").
Moderating variable: changes the strength/direction of the IV–DV relationship (e.g., "market competitiveness").
Mediating variable: explains how or why the IV affects the DV (e.g., "brand awareness" mediating advertising → sales).

✅ Likely Exam Questions — Unit 2

1. Define literature review and state its objectives.
2. Differentiate primary, secondary, and tertiary sources with examples.
3. Explain Boolean operators (AND/OR/NOT) with an example search string.
4. Compare any three types of literature review (e.g., systematic vs integrative vs critical).
5. What is a research gap? How is it identified?
6. Differentiate a research problem from a research objective.
7. Write a general and a specific (SMART) research objective for a given topic.
8. Write a null and an alternative hypothesis for a given research scenario.
 Unit 3 — Research Design
The blueprint for collecting, measuring, and analyzing data

3.1 Definition
Research Design is the overall strategy chosen to integrate the different components of a study in a coherent and logical way, so as to
effectively address the research problem. It is the blueprint for the collection, measurement, and analysis of data.

3.2 Types of Research Design

Type Definition Example

Exploratory Used when the problem is not clearly defined; gathers preliminary information to Informal interviews/focus groups to explore why
understand issues. "Let's explore first before we fix or decide anything." extracurricular participation is falling

Descriptive Describes the characteristics of a population or phenomenon; answers who, what, when, Survey on % of students preferring online vs
where, how — not why face-to-face classes

Experimental Tests cause-and-effect by manipulating variables under controlled conditions; one group Comparing English proficiency after traditional
gets treatment, another doesn't, and you compare lectures vs interactive apps

Causal Finds cause-and-effect relationships in a real-world (less controlled) setting — similar to Studying whether career counseling increased
experimental but happens outside the lab internship placements

Which Design Fits Which MBA Specialization?

Specialization Suitable Design(s) Why

HRM Descriptive, Exploratory, Employee behavior/job satisfaction can start with exploration, be measured descriptively, or studied via causal
Causal links (motivation vs performance)

Finance Causal, Experimental Examines cause-effect (interest rates → investment; financial literacy → savings); experiments used in
behavioral finance

Management Exploratory, Descriptive, Explore leadership styles, describe management trends, examine causes of organizational success
Causal

Supply Chain Descriptive, Causal, Descriptive maps logistics/inventory patterns; causal/experimental study technology's impact on efficiency
Mgmt Experimental

Quick Student Tips

Exploratory = new or under-studied areas · Descriptive = perfect for survey-based studies · Causal/Experimental = best when testing
"Does X cause Y?"

✅ Likely Exam Questions — Unit 3

1. Define research design and explain its role in a study.
2. Differentiate the four types of research design with examples.
3. Which research design would you use to study "the effect of remote work on employee productivity"? Justify.
4. Compare experimental and causal research designs.
5. Recommend a suitable research design for a Finance-specialization MBA thesis and justify your choice.
 Unit 4 — Sampling Techniques & Sample Size Determination
Choosing who to study, and how many

4.1 What is Sampling?

Term Definition Example

Population Entire group the researcher wants to study All customers of a retail chain

Sample A smaller group selected from the population 500 customers randomly selected for a survey

Sampling is the process of selecting a subset of individuals/units from a larger population to participate in a study.

Why sample? Populations can be too large/inaccessible; sampling saves time, money, effort; proper sampling allows generalization to the
whole population.

4.2 Probability Sampling Techniques ✅

Rely on random selection — each member has a known, equal chance of being chosen. Essential for generalizable results.

Type How It Works Example

Simple Every member has an equal chance (like drawing names from a hat) Software randomly selects 300 of 10,000 customers for a
Random survey

Systematic Select every kth member from a list after a random start From 1,000 employees, select every 10th starting from a
random number between 1–10

Stratified Population divided into strata (subgroups) by key characteristics; random Bank samples 100 customers randomly from each of
samples drawn proportionally from each North/South/East/West regions

Cluster Population divided into clusters (cities, branches); some clusters randomly Retailer surveys all customers in 5 randomly selected stores
selected, then sampled fully or partly out of 50

Systematic sampling interval: k = Population Size ÷ Sample Size

Example: Population = 1,000, desired sample = 100 → k = 1000/100 = 10 → pick every 10th person after a random
start between 1–10.
⚠ Caution: if the list has a hidden pattern repeating every k members, the sample can become biased — always
check for patterns first.

Method Pros Cons

Simple Random Easy, unbiased Requires a complete list

Systematic Simple, fast Can be biased if a pattern exists

Stratified Ensures subgroup representation Requires subgroup information

Cluster Cost-effective for spread-out populations Less precise than stratified

4.3 Non-Probability Sampling Techniques ❌

Do not rely on random selection. Useful for exploratory/qualitative research but limit generalizability.

Type How It Works Example

Convenience Select the most easily accessible participants Surveying MBA classmates because they're easy to reach

Purposive (Judgmental) Researcher selects participants based on specific characteristics Interviewing only marketing managers about branding strategy

Snowball Existing participants recruit new participants from their network Studying startup founders via referrals from initial contacts

Quota Researcher sets subgroup quotas, selects non-randomly until filled Interview 50 males and 50 females from a customer database

Method Pros Cons

Convenience Fast, low cost High bias, non-representative

Purposive Focused, expert insights Subjective, limited scope

Snowball Access to hard-to-reach populations Bias, non-random

Quota Ensures subgroup numbers Non-random, possible bias

4.4 Sample Size Determination

Why it matters: affects accuracy, reliability, generalizability. Too small → low power, unreliable results. Too large → wasted resources.

Method Description Example

Rules of Thumb Simple general guidelines when exact calculations aren't Use 30+ respondents per group (3 groups → 90–100 total)
available

Power Analysis Statistical method using effect size, significance level (α), and Detecting a 10% sales improvement at 80% power might need ~200
desired power respondents

Effect Size Larger expected effects need fewer participants; smaller effects Detecting a small behavior change needs a bigger sample than
Consideration need more detecting a big one

Exam-relevant addition — Sample size formulas (not shown in slides, but standard exam material)

Slovin's Formula (simple, widely used when population is known and finite):
n = N / (1 + N·e²)
where n = sample size, N = population size, e = margin of error (as a decimal, e.g., 0.05 for 5%)

Example: N = 1,000, e = 0.05 → n = 1000 / (1 + 1000×0.0025) = 1000 / 3.5 ≈ 286

Cochran's Formula (for large/unknown populations, proportions):

n₀ = Z²·p·q / e²
where Z = Z-score for confidence level (1.96 for 95%), p = estimated proportion (0.5 if unknown, maximizes
sample size), q = 1 − p, e = margin of error

Example: Z=1.96, p=0.5, q=0.5, e=0.05 → n₀ = (1.96²×0.5×0.5)/0.05² ≈ 384

Worked Example — Power Analysis

Parameter Value

Significance level (α) 0.05 (5% chance of false positive)

Power 80% (chance of detecting a true effect)

Expected Effect Size (Cohen's d) Moderate (0.5)

Result: required sample size per group ≈ 64 employees (i.e., at least 64 in control and 64 in the training group) to detect a meaningful
productivity difference.

4.5 Alpha (α) and Beta (β)

Error
Term Meaning Example
Type

Alpha Probability of a false positive — concluding an effect exists when it doesn't. Usually set Type I Saying a new ad works when it actually
(α) at 0.05 error doesn't

Beta Probability of a false negative — failing to detect a real effect. Power = 1 − β Type II Saying a new ad doesn't work when it
(β) (commonly 80–90%) error actually does

Analogy: a smoke detector — Alpha = false alarm (screams when there's no fire); Beta = missed detection (stays silent during a real fire).

4.6 Challenges in Sampling & Practical Solutions

Challenge Description Solution

Access to Population Hard to find/reach the right people Snowball sampling (referrals) or purposive sampling (target specific people)

Sampling Bias Sample doesn't represent the population Randomization or stratified sampling

Non-response Bias Selected people refuse/ignore surveys Follow-up reminders, incentives, short/easy surveys

Time & Cost Constraints Limited budget/time Balance sample size vs precision; use online surveys

Real-World Sampling Examples

Scenario Method Why

Customer satisfaction across regions Stratified Ensures every region is represented fairly

Exploring startup challenges Snowball Startups are hard-to-reach

Employee engagement survey Simple random Generalizable company-wide results

 Quick feedback on a new product Convenience Speed/ease over representativeness

Recommended Resources
Research Methods for Business Students (Saunders, Lewis & Thornhill); Sampling Design and Analysis (Sharon Lohr); Tools: G*Power (power analysis), SPSS
Sample Power, Qualtrics/SurveyMonkey.

✅ Likely Exam Questions — Unit 4

1. Differentiate probability and non-probability sampling with two examples of each.
2. A list has 1,200 employees and you need a sample of 120. Calculate k and describe the systematic sampling procedure.
3. Explain stratified vs cluster sampling — when would you use each?
4. Define Type I and Type II errors, and relate them to alpha and beta.
5. Using Slovin's formula, calculate the sample size for a population of 500 at a 5% margin of error.
6. What challenges arise in sampling and how can a researcher address non-response bias?
7. Explain why sample size affects the reliability of research findings.
 Unit 5 — Data Collection Methods & Measurement/Scaling
Techniques
Part 1: Collecting data · Part 2: Measuring it accurately

5.1 Primary vs Secondary Data

Feature Primary Data Secondary Data

Source Collected by the researcher directly Collected by someone else (institution)

Purpose For the specific study at hand Originally for another purpose

Time & Cost Time-consuming, expensive Time-saving, low-cost

Relevance Highly relevant, specific May be partially relevant/outdated

Control Full control over methodology/accuracy Limited control over collection process

Examples Surveys, experiments, interviews Academic articles, company reports, databases

Benefits Limitations

Primary Accurate/reliable if collected properly; specific to the question; more Expensive, time-intensive; needs skilled personnel; possible
control over quality ethical/privacy constraints

Secondary Accessible, low-cost; historical context; useful for exploratory phases May be outdated/irrelevant; quality uncertain; little info on how it was
collected

When to Use Which?

Scenario Recommended

Early-stage exploration Secondary

Testing a new hypothesis Primary

Lack of budget/time Secondary

Detailed customer feedback needed Primary

5.2 Qualitative vs Quantitative Data

Feature Qualitative Quantitative

Data Type Text, audio, video Numbers, percentages

Objective Explore phenomena Measure and quantify variables

Tools Interviews, focus groups Surveys, experiments

Analysis Method Thematic coding Statistical analysis

Qualitative: a focus group discussing consumer behavior. Quantitative: a survey rating customer satisfaction from 1–5.

5.3 Survey, Interview & Focus Group Methods

Surveys: structured questionnaires, closed-ended questions; online/phone/in-person. E.g., customer satisfaction survey with Likert-
scale ratings.
Interviews: one-on-one, in-depth conversations; structured, semi-structured, or unstructured. E.g., interviewing business owners
about economic downturns.
Focus Groups: 6–10 participants, moderated discussion; useful for diverse perspectives. E.g., testing marketing messages for a new
campaign.

5.4 Observation, Case Study & Document Analysis

Observation: watching/recording behavior in natural settings; participant or non-participant. E.g., observing customer flow in a
retail store.
Case Study: in-depth analysis of a single unit (org/event/individual), using multiple data sources. E.g., a company's digital
transformation journey.
Document Analysis: analyzing existing documents (reports, minutes, memos) for historical/contextual information. E.g., reviewing
annual reports for CSR trends over 5 years.
 5.5 Measurement Scales (NOIR)
Measurement scales are the rules used to assign numbers/labels to variables — they determine which statistical tests can be used.

Scale Description Example Key Characteristic

Nominal Categorized without order Gender, Department (HR, IT) No order, no numerical meaning

Ordinal Ordered categories, unequal intervals Education level (High School < Bachelor's < Master's) Rank matters, distance doesn't

Interval Ordered, equal intervals, no true zero Temperature (°C), IQ scores Can add/subtract, not multiply/divide

Ratio Like interval, but has a true zero Income, Age, Sales, Weight All mathematical operations valid

Why it matters: Nominal & Ordinal → non-parametric tests. Interval & Ratio → allow correlation, regression, and other parametric tests. Note: 0°C is not "no
temperature" — it's just a point on the scale (no true zero).

Memory aid
The acronym NOIR (Nominal → Ordinal → Interval → Ratio) is the standard way this scale hierarchy is remembered in exams — each
level up adds more mathematical properties than the one before.

5.6 Reliability & Validity of Measurement Instruments

Reliability — consistency/stability of an instrument

Test–Retest Reliability: same instrument, same people, two time points → similar results
Inter-Rater Reliability: different observers score the same phenomenon consistently
Internal Consistency: items measuring the same concept correlate well; commonly assessed with Cronbach's Alpha

A satisfaction survey run in March and again in April giving similar results shows good test–retest reliability.

Exam-relevant addition — Cronbach's Alpha threshold (not stated numerically in slides)

Cronbach's Alpha ranges from 0 to 1. A value of ≥ 0.70 is generally considered acceptable internal consistency; ≥ 0.80 is good; ≥
0.90 is excellent (though very high values can indicate redundant items).

Validity — accuracy/truthfulness of a measurement

Content Validity: instrument covers all relevant aspects of the concept. E.g., a leadership questionnaire should include
communication, decision-making, etc.
Construct Validity: instrument actually measures the theoretical construct intended. E.g., a motivation scale measuring intrinsic vs
extrinsic motivation.
Criterion Validity: compares the instrument against an external benchmark. E.g., comparing test scores with final course grades.

An employee-engagement scale should correlate with real engagement behaviors (attendance, punctuality, productivity).

5.7 Developing Questionnaires & Scales

Design Steps
1. Define the objective — what exactly are you measuring?
2. Choose question types — open-ended (detailed but harder to analyze) vs closed-ended (easier to analyze, e.g., multiple
choice/Likert)
3. Use clear, neutral wording — avoid jargon and leading questions
4. Pre-test the questionnaire (pilot study) on a small group
5. Revise and refine using feedback; check reliability and validity before final use

Best Practices
Use Likert Scales (e.g., 1 = Strongly Disagree to 5 = Strongly Agree) for attitudes/perceptions
Avoid double-barreled questions — ❌ "Do you like our product quality and pricing?" → ✅ split into two separate questions
Maintain logical flow — group similar topics, use section headings
Ensure clear instructions on how to complete the questionnaire

✅ Likely Exam Questions — Unit 5

1. Differentiate primary and secondary data with examples and state when each is preferable.
2. Differentiate qualitative and quantitative data collection.
3. Compare surveys, interviews, and focus groups as data collection tools.
4. Explain the four measurement scales (NOIR) with one example each.
5. Differentiate reliability and validity; name the types of each.
6. What is a double-barreled question? Give an example and correct it.
7. List the steps in designing a questionnaire.
 Unit 6 — Modern Data Collection Tools & Techniques
Digital transformation of business research data

6.1 Why Modern Tools Matter

Digital transformation has created new data sources
More accessible, real-time, and scalable than traditional methods
Enhanced accuracy and broader reach

6.2 Online Surveys & Mobile Data Collection

Digital forms/apps have largely replaced paper surveys — enabling wider reach, real-time analysis, and cost-effective scalability.

Tool Platform Features

Qualtrics Web/Email Advanced logic, branching, export options

SurveyMonkey Web/Mobile Templates, real-time results, analytics

Google Forms Free/Cloud Easy to use, ideal for quick data collection

KoBoToolbox Mobile/Offline Best for remote or field data collection (e.g., NGOs); syncs later without internet

6.3 Social Media Analytics & Sentiment Analysis

Platforms like Twitter, Facebook, Instagram give real-time insight into public opinion, trends, customer preferences, and brand perception.

TextBlob, VADER — Python libraries for sentiment analysis

Brandwatch, Hootsuite — dashboards for monitoring social media activity/sentiment
AI Integration — detects emotions, recurring keywords/hashtags

A company analyzes tweets during a product launch to gauge reactions and respond quickly.

6.4 Web Scraping & Big Data Tools

Web scraping automates data collection from websites/forums/reviews; big data tools manage and process vast unstructured datasets.

Tools: BeautifulSoup, Scrapy (Python scraping); Google BigQuery, AWS, Hadoop (storage/processing)
Ethics: always check a website's [Link]; comply with data-privacy laws (GDPR, etc.)

A hotel chain scrapes travel-site reviews to find recurring complaints and improve service.

6.5 IoT Devices & Sensor-Based Data

Internet of Things (IoT) devices collect continuous, real-time data from the physical world — used in logistics, healthcare, agriculture,
smart homes.

Logistics: GPS trackers for route optimization

Healthcare: wearables tracking heart rate, sleep
Agriculture: soil-moisture sensors guiding irrigation

Challenges: high data volume/complexity; privacy and cybersecurity concerns.

6.6 Case Study — Real-Time Marketing with Social Listening

A beverage company monitors social media during a high-profile event (e.g., Super Bowl) to gauge reactions to its ad campaign —
using Brandwatch to track mentions/sentiment and AI to classify tweets as positive/neutral/negative. Result: real-time response to
feedback and adjusted strategy based on trends. Key insight: listening to real-time data lets brands be proactive rather than
reactive.

6.7 Traditional vs Modern Tools

Feature Traditional Modern

Speed Slow, manual Real-time, automated

Scale Local, limited Global, massive reach

Accuracy Prone to manual error More consistent, tech-assisted

 Cost Higher (manpower) Lower per response

Engagement Often passive Interactive, user-friendly

Modern tools enable more meaningful, continuous engagement — but require technical skill and good data governance.

6.8 Best Practices for Using Modern Tools

Choose tools based on research/business objectives
Verify data quality and source credibility
Ensure compliance with data protection laws (GDPR, HIPAA)
Combine multiple tools (e.g., web scraping + sentiment analysis) for richer insight
Stay updated — today's tools can become outdated quickly

6.9 Data Protection Laws — Quick Reference

Law Region Covers Status in Pakistan

GDPR EU & Global All personal data Applies only if handling EU citizens' data

HIPAA USA Health/medical data Not applicable

PDPB Pakistan All personal data in Pakistan Drafted, not fully enforced yet (final-draft stage 2024–25)

PECA 2016 Pakistan Cybercrimes, unauthorized data access Enforced (limited scope)

GDPR requires explicit consent and mandates breach reporting within 72 hours; grants individuals rights to access/correct/delete data. HIPAA covers only U.S.
health-related information. Until Pakistan's PDPB is enacted, companies/researchers should still: get informed consent, avoid over-collecting data, store data
securely and anonymize where possible, and never sell/share personal data without permission.

✅ Likely Exam Questions — Unit 6

1. List and explain four modern data collection tools/techniques.
2. Compare traditional and modern data collection tools on speed, scale, accuracy, cost, and engagement.
3. What ethical precautions should be taken when web scraping?
4. Explain GDPR, HIPAA, and PDPB — what each covers and their status in Pakistan.
5. Using the case study, explain how social listening supports real-time marketing decisions.
 Unit 7 — Ethical Considerations in Modern Data Collection
Responsible research in the age of AI, social media, and big data

Modern data collection tools offer unparalleled speed, scale, and insight — but raise serious ethical challenges. It is the researcher's
responsibility to ensure data is collected, stored, and analyzed in ways that respect participants' rights and privacy, especially where
formal regulation (as in Pakistan) is still evolving.

7.1 Key Areas to Cover in a Thesis/Report

Section Key Focus

Introduction Role of data in modern business; why ethics matter

Literature Review Studies on ethical practices in digital research

Modern Tools Overview Online surveys, social media mining, AI tools, web scraping

Ethical Considerations Informed consent, data ownership, privacy, bias

Pakistan's Legal & Ethical Context PDPB, PECA, cultural expectations, corporate responsibility

Recommendations Best practices for ethical MBA-level research

Conclusion Summary of findings, business implications

7.2 Six Core Ethical Principles

# Principle Meaning Modern Challenge Best Practice

1 Informed Participants understand what data Website/social-media data often collected Clear consent forms/pop-ups/disclaimers;
Consent is collected and how it's used without user knowledge consider expected privacy even for public data

2 Privacy & Data stays confidential, used only Cloud tools/shared servers raise leak risk Anonymization, encrypted storage, limited
Confidentiality for the stated purpose access

3 Data Ownership Participants know who owns their AI/platforms may reuse data without explicit Be transparent on storage duration,
& Usage Rights data and how it's shared/stored permission destination, access

4 Bias & Fairness Data fairly represents all AI/big data can amplify existing bias in Test tools for bias; strive for diverse,
groups/perspectives unbalanced datasets representative samples

5 Transparency & Methods are clearly explained; AI "black box" — decisions not easily explained Keep a clear audit trail of data collection,
Accountability researchers take responsibility analysis, interpretation

6 Legal Follow applicable law Pakistan's PDPB still in development; PECA Even absent strict laws, follow international
Compliance 2016 criminalizes unauthorized data access ethical standards (e.g., GDPR)

Self-Check Questions for Researchers

Did the participant agree to provide this data? · Is the data secure and anonymized? · Who owns the data, and how will it be
used? · ⚖ Does my tool/method disadvantage any group? · Are my methods and intentions clear?

7.3 Pakistan's Legal & Ethical Framework

Personal Data Protection Bill (PDPB) — drafted, inspired by GDPR, not yet enforced; would apply to any entity
collecting/processing/storing personal data of individuals in Pakistan and proposes a National Commission for Personal Data
Protection (NCPDP)
PECA 2016 (Prevention of Electronic Crimes Act) — criminalizes unauthorized access to personal data (enforced, limited scope)
Partial coverage also comes from PTA telecom policies and SECP regulations for financial institutions
Until PDPB is fully enacted: follow institutional ethics review procedures and global best practices (e.g., GDPR)

7.4 Traditional vs Modern Tools — Ethical Comparison

Criteria Traditional Methods Modern Tools

Consent Usually written Often unclear or missing

Data Volume Limited Massive (big data, real-time)

Risk of Exposure Lower Higher (cloud, scraping, AI)

Anonymity Easier Requires technical processing

Ethics Awareness Higher Often neglected in automation

 7.5 Writing Ethics into a Thesis/Report
What to include:

A data privacy and confidentiality section

A statement on informed consent
Explanation of data sources and whether they're public
Limitations and potential risks or bias
Mention of compliance with institutional/national ethics guidelines

Example Ethics Statement

"This study used publicly available data from Twitter. No personal identifiers were collected. Ethical approval was obtained from XYZ
University's review committee."

Tips for Ethical Digital Research

Ask: would I be comfortable if my own data were used this way?
Avoid tools/shortcuts that violate terms of service
Don't collect more data than needed
Anonymize and encrypt sensitive data
Cite your tools and explain how they work
Keep an "Ethical Research Checklist" for every project

✅ Likely Exam Questions — Unit 7

1. List and explain the six core ethical principles in modern data collection.
2. What is informed consent, and why is it harder to obtain with modern digital tools?
3. Explain Pakistan's current legal framework for data protection (PDPB, PECA).
4. Compare the ethical risks of traditional vs modern data collection tools.
5. What should a researcher include in the "ethics" section of a thesis? Give an example statement.
 Appendix
Quick-reference material for last-minute revision

A.1 Thesis Chapter Roadmap

How these lecture topics typically map onto an MBA thesis structure:

Topic Typical Thesis Chapter

Research Design and Planning Chapter 1: Introduction

Literature Review Chapter 2: Literature Review

Data Collection & Measurement Techniques Chapter 3: Methodology (Part 1)

Sampling Techniques & Sample Size Chapter 3: Methodology (Part 2)

Data Collection Tools & Techniques Chapter 3: Methodology (Part 3)

Measurement and Scaling Chapter 3: Methodology (Part 4)

Data Analysis Techniques Chapter 4: Data Analysis

Ethical Considerations Chapter 5: Ethics & Limitations

Writing and Presenting Findings Chapter 6: Conclusion & Reporting

A.2 Master Formula Sheet

Systematic sampling interval: k = Population Size ÷ Sample Size

Slovin's Formula (finite population): n = N / (1 + N·e²)

Cochran's Formula (large/unknown population): n₀ = Z²·p·q / e² (use p = q = 0.5 if unknown, Z = 1.96 for 95%
confidence)

Power: 1 − β (β = Type II error rate; commonly targeted at 80–90% power)

Common significance level (α): 0.05 (5%)

Cronbach's Alpha acceptable threshold: ≥ 0.70

A.3 Master Glossary

Research
Systematic, organized investigation of a problem to reach a solution.
Hypothesis
A testable statement predicting a relationship between variables.

Validity
Whether an instrument measures what it claims to measure.

Reliability
Whether an instrument produces consistent results over repeated use.
Population
The entire group a researcher wants to study.
Sample
A subset of the population selected for study.
Probability Sampling
Random selection giving every unit a known chance of selection (simple random, systematic, stratified, cluster).
Non-Probability Sampling
Non-random selection (convenience, purposive, snowball, quota).

Type I Error (α)

False positive — concluding an effect exists when it does not.
Type II Error (β)
False negative — failing to detect a real effect.
Nominal / Ordinal / Interval / Ratio (NOIR)
The four levels of measurement scale, in increasing order of mathematical properties.
Literature Review
Critical synthesis of existing research on a topic to identify what is known and what gaps remain.
Research Gap
An area of missing, conflicting, or outdated information in existing literature.
Cronbach's Alpha
A statistic measuring internal consistency reliability of a set of scale items.
GDPR / HIPAA / PDPB
Data-protection regulations of the EU, USA (health data), and Pakistan (draft), respectively.

A.4 High-Yield Cross-Topic Exam Questions

✅ Comprehensive / Integrative Questions

1. Trace the full research process from problem identification to reporting, mapping each step to the corresponding thesis chapter.
2. A company wants to know why sales dropped in a region. Recommend a research design, sampling method, sample size
approach, and data collection tool, and justify each choice.
3. Explain how validity, reliability, and measurement scale choice all affect the credibility of survey-based research.
4. Compare the ethical obligations of a researcher using a traditional paper survey vs. one scraping social media data.
5. Design a mini research proposal (problem, objective, hypothesis, design, sampling method, sample size, data collection tool, and
one ethical consideration) for a topic of your choice.

Compiled from course lecture slides (Aleena Khan) for exam revision. Where lecture content did not explicitly cover a standard topic, additional material has been added and
clearly marked with a tag throughout this guide.