Research Methodology Complete Solutions

COMPREHENSIVE ANSWER BANK • MCA 2ND SEM (PATTERN 2024)

PART A: SHORT ANSWER / CONCEPTUAL QUESTIONS

1. Define Research.
Research is a structured, systematic, and objective investigation aimed at discovering new facts, verifying
existing knowledge, establishing relationships, and reaching novel conclusions. It follows a scientific method
to solve problems, close knowledge gaps, and expand the horizons of a discipline through purposeful data
collection, critical analysis, and logical interpretation.

2. What is the significance of primary data?

Primary data is highly significant because it is first-hand information collected directly from the source for a
specific research objective. Its primary advantages include:
• High Reliability & Specificity: It is customized to address the exact objectives and questions of the study.
• Ownership and Authenticity: The researcher controls data collection parameters, ensuring unparalleled
data integrity and control over biases.
• Recency: It captures real-time, current realities and up-to-date trends that secondary data sources might
lack.

3. What is Exploratory Research?

Exploratory research is a preliminary study conducted when a problem is vague, poorly understood, or ill-
defined. It does not aim to provide definitive answers or conclusive evidence; instead, it seeks to understand
the nature of the problem, formulate precise hypotheses, operationalize variables, and determine the
feasibility of a full-scale, conclusive study. It relies heavily on qualitative techniques like literature reviews, pilot
interviews, case studies, and focus group discussions.

4. Define Research Design.

A Research Design is the conceptual structure or blueprint within which research is executed. It constitutes
the formal plan for data collection, measurement, and analysis. It outlines the strategic framework specifying
what data to collect, from which sources, via what methods, and how the gathered facts will be analyzed to
answer the research questions efficiently while minimizing expenditure of time, effort, and money.

MCA 2nd Sem (Pattern 2024) — Research Methodology Reference Guide 1

5. What is a Hypothesis? Define two types of hypothesis.
A Hypothesis is a tentative, testable proposition or statement about the expected relationship between two or
more variables. It serves as a guide for empirical testing.
The two core types of hypotheses are:
• Null Hypothesis (H ): A statement claiming that there is no statistical significance, no actual effect, or no
0
real difference between the studied groups or variables. Any observed variance is attributed to random
chance.
• Alternative Hypothesis (H or H ): A statement that runs counter to the null hypothesis, asserting that a
1 a
real effect, pattern, connection, or statistically significant difference exists between the variables.

6. What is a Likert Scale?

A Likert Scale is a psychometric, closed-ended rating scale used extensively in questionnaires to evaluate
respondents' attitudes, opinions, or levels of agreement regarding a specific statement. It typically provides a
symmetrical continuum of 5 or 7 options ranging from one extreme to another (e.g., Strongly Disagree,
Disagree, Neutral, Agree, Strongly Agree). Responses are assigned numerical weights to calculate aggregate
scores for metric analysis.

7. Differentiate between Probability and Non-Probability Sampling.

The fundamental distinction lies in how sample elements are selected:
• Probability Sampling: Every element in the target population has a known, non-zero, and equal chance
of being selected. It uses randomization techniques, eliminates selection bias, and allows for statistical
generalization.
• Non-Probability Sampling: Selection relies on the subjective judgment, convenience, or deliberate
choice of the researcher. Elements do not have an equal chance of selection, making it impossible to
compute sampling error or generalize results statistically to the broader population.

8. What is the significance of secondary data?

Secondary data consists of information collected and compiled by someone else in the past for another
purpose. Its significance includes:
• Resource Efficiency: Saves substantial time, effort, and capital compared to primary data collection.
• Context and Foundation: Helps establish historical context, frames the problem, and benchmarks new
primary findings.
• Access to Scale: Grants access to massive datasets (e.g., national census reports, large government
archives) that are practically impossible for an individual researcher to gather.

9. What is the Concept of Measurement?

In research, Measurement is the systematic process of assigning numbers, symbols, or values to properties
of objects, individuals, or phenomena according to explicitly predefined, rule-based criteria. It translates

MCA 2nd Sem (Pattern 2024) — Research Methodology Reference Guide 2

abstract, qualitative concepts (such as customer satisfaction or brand loyalty) or concrete entities into precise
data points that can be systematically mapped, compared, and analyzed.

10. Differentiate between Nominal and Ordinal Scale with examples.

• Nominal Scale: The simplest measurement scale where numbers or labels serve purely as identifiers or
category names. There is no intrinsic ordering, ranking, or quantitative value implied.

Example: Gender (1 = Male, 2 = Female), Software Development Roles (QA, Developer, Architect), or eye
color.

• Ordinal Scale: A scale that groups items into discrete categories that possess a clear, logical progression,
rank, or order. However, the exact mathematical difference between ranks is unknown or unequal.

Example: Project Priority (High, Medium, Low), Student Grade (A, B, C), or Likert agreement rankings.

11. What is Interval Scale and Ratio Scale? Explain with examples.
• Interval Scale: A scale with ordered categories where the distance (interval) between any two
consecutive points is completely equal and measurable. It does not possess an absolute or true zero
point; zero is assigned arbitrarily.

Example: Temperature in Celsius or Fahrenheit. The difference between 20°C and 30°C equals that
between 30°C and 40°C. However, 0°C does not mean an absence of heat.

• Ratio Scale: The most advanced scale, possessing all characteristics of an interval scale along with a
true, absolute zero point representing the total absence of the measured attribute. This allows for
meaningful multiplicative and ratio statements.

Example: Monthly Income, Execution Time (seconds), or Weight. An execution time of 0 seconds means
zero duration, and a 10-second process takes exactly twice as long as a 5-second process.

MCA 2nd Sem (Pattern 2024) — Research Methodology Reference Guide 3

 PART B: DIFFERENTIATION & COMPARISON BASED QUESTIONS

12. Mean vs. Median - Differentiate with suitable example.

Feature Mean (Arithmetic Average) Median (Middle Value)

Definition The sum of all values divided by the The exact physical midpoint of a
total count of observations. dataset sorted in ascending or
descending order.

Sensitivity to Outliers Highly sensitive; extreme values distort Highly robust; extreme values do not
the mean significantly. shift the center position.

Mathematical Base Uses all data points (Σx / n). Positional measure (identifies the
(n+1)/2 th position).

Example Scenario: Consider software developer salaries in a startup (in thousands): 40, 45, 50, 55, 300
(where 300 belongs to the founder).
• Mean: (40 + 45 + 50 + 55 + 300) / 5 = 490 / 5 = 98. The mean salary is $98k, which doesn't accurately
represent the typical developer due to the single outlier.
• Median: Sorted array: [40, 45, 50, 55, 300]. The middle value is $50k. This provides a much fairer
representation of a developer's typical earnings.

13. Descriptive vs. Exploratory Research - Differentiate.

Criteria Exploratory Research Descriptive Research

Core Objective To discover insights, formulate To describe the characteristics, behaviors,

hypotheses, and clarify an ambiguous or profiles of a population or
problem. phenomenon.

Structure Highly flexible, unstructured, informal, and Rigid, highly structured, structured
iterative. instruments, and pre-planned.

Hypothesis Formulated during or at the end of the Starts with a clear hypothesis or specific
research process. research questions.

Data Focus Predominantly qualitative (interviews, Quantitative and structured (surveys,

open discussions). structured observation).

MCA 2nd Sem (Pattern 2024) — Research Methodology Reference Guide 4

14. Descriptive vs. Experimental Research - Differentiate.

Criteria Descriptive Research Experimental Research

Variable Control No control over variables. The researcher Deliberate control and manipulation of
observes and measures things as they independent variables to observe their
naturally occur. effects.

Causality Establishes associations, correlations, Establishes definitive cause-and-effect

patterns, or profiles, but cannot prove relationships via rigorous control groups.
cause-and-effect.

Setting Conducted in natural environments (field Conducted in controlled artificial

studies, census, market surveys). environments (laboratories) or structured
fields.

15. Rating Scale vs. Ranking Scale - Differentiate with examples.

Feature Rating Scale Ranking Scale

Core Evaluation Assesses each item independently on an Compares items relative to one another to
absolute continuum or score metric. establish an ordinal preference list.

Ties Multiple objects can receive identical Ties are typically prohibited or restricted;
scores (e.g., three items rated 5 stars). forces distinct placement ordering (1st,
2nd, 3rd).

Cognitive Load Low; simple to score each criterion High; requires respondents to weigh and
individually. balance all options collectively.

Examples:
• Rating: Rate 3 programming languages on a scale of 1-5 for ease of use. (Python: 5/5, Java: 4/5, C++:
3/5).
• Ranking: Rank the following languages in order of preference from 1 to 3: Python, Java, C++. (1st: Python,
2nd: Java, 3rd: C++).

MCA 2nd Sem (Pattern 2024) — Research Methodology Reference Guide 5

16. Univariate vs. Bivariate Analysis - Differentiate.

Criteria Univariate Analysis Bivariate Analysis

Number of Examines exactly one variable at a time. Examines two variables simultaneously.
Variables

Analytical To describe, summarize, and find patterns To discover relationships, correlations,

Purpose within a single distribution. dependencies, or associations between
two distinct factors.

Statistical Tools Frequency distributions, Mean, Median, Scatter plots, Pearson correlation
Mode, Variance, Standard Deviation, coefficient (r), Simple Linear Regression,
Histograms. Chi-Square tests.

MCA 2nd Sem (Pattern 2024) — Research Methodology Reference Guide 6

 PART C: LONG ANSWER / DESCRIPTIVE QUESTIONS

17. Explain types of Probability Sampling with suitable examples.

Probability sampling operates on mathematical randomization, ensuring each member has a known chance
of selection. The core types are:
1. Simple Random Sampling (SRS): Every element has an equal probability of selection. Drawn using
automated random number generators or lottery systems.

Example: Selecting 50 student IDs out of a database of 500 using an Excel random number generator
script.

2. Stratified Random Sampling: The population is segregated into mutually exclusive, homogeneous
subgroups (strata) based on specific characteristics (e.g., age, income, stream). A simple random sample
is then drawn from each stratum.

Example: In an IT company survey, dividing employees into strata of Developers, HR, and Managers, then
drawing proportional samples from each group to capture all organizational perspectives.

3. Systematic Sampling: Elements are selected from an ordered sampling frame at fixed, regular geometric
intervals determined by a sampling interval k = N / n (where N is population size and n is sample size),
starting from a randomly chosen point.

Example: Selecting every 10th customer entering a retail store, beginning with a randomly selected 3rd
customer.

4. Cluster Sampling: The population is partitioned into heterogeneous, geographically diverse blocks or
groups called clusters. The researcher randomly selects entire clusters, and then surveys all elements
within those chosen clusters.

Example: To study banking habits in a state, a researcher randomly selects 5 specific districts (clusters)
and studies all residents within those districts.

18. Explain types of Non-Probability Sampling with suitable examples.

Non-probability sampling relies on non-random, human-centric criteria. The core types are:
1. Convenience Sampling: The sample is selected based entirely on ease of access, physical proximity, or
availability to the researcher.

Example: A college researcher interviewing fellow students standing outside the campus library.

2. Purposive / Judgmental Sampling: The researcher uses specialized knowledge and subjective
discretion to handpick individuals who best fit the research profile or objectives.

MCA 2nd Sem (Pattern 2024) — Research Methodology Reference Guide 7

 Example: To study the impact of a rare cloud infrastructure bug, a researcher purposefully selects cloud
architects with over 15 years of infrastructure experience.

3. Quota Sampling: The researcher sets explicit targets or quotas for specific subgroups (e.g., 60% male,
40% female) to mirror population distributions, but selects the actual respondents within those quotas non-
randomly.

Example: Stopping shoppers in a mall until exactly 50 iPhone users and 50 Android users have completed
the survey questionnaire.

4. Snowball (Network) Sampling: Used when target populations are hard to reach or hidden. Initial
respondents are located and, after being interviewed, are asked to refer others from their social network
who meet the criteria.

Example: Researching the challenges faced by underground, undocumented gig-workers or elite hackers.

19. Explain Sampling Errors and Non-Sampling Errors in detail.

Errors in research degrade data validity. They are broadly categorized into two types:

A. Sampling Errors
These occur purely due to the nature of chance when taking a sample instead of conducting a full census. It
represents the divergence between the true population parameter and the sample statistic.
• Characteristics: It is unavoidable unless the whole population is measured. It decreases systematically
as the sample size increases.
• Causes: Bad sample design, chance selection of non-representative elements.

B. Non-Sampling Errors
These errors stem from human, operational, or systemic failures during research execution, data collection, or
data processing. They can occur in both sample surveys and complete censuses.
• Characteristics: They do not decrease with larger sample sizes; in fact, they can grow worse if field
teams are poorly trained.
• Key Categories:
◦ Data Acquisition Errors: Poorly framed, ambiguous questions that confuse respondents, or leading
questions that introduce bias.
◦ Non-Response Error: Occurs when selected individuals refuse to participate or cannot be reached,
skewing results if non-respondents share distinct traits.
◦ Response/Measurement Bias: Respondents deliberately giving false, socially acceptable answers or
recording tools failing.
◦ Processing Errors: Mistakes during manual data entry, coding, tabulation, or software execution.

MCA 2nd Sem (Pattern 2024) — Research Methodology Reference Guide 8

20. Explain the Structure of a Research Report.
A formal research report follows a standard sequence to ensure clarity, replicability, and academic rigour:
1. Preliminary Pages:
◦ Title Page: Precise title, author details, institutional affiliation, and date.
◦ Abstract / Executive Summary: A concise synopsis (~200-300 words) summarizing the problem
statement, methodology, primary findings, and conclusions.
◦ Table of Contents & Lists: Hierarchical map of chapters, figures, and tables.
2. Main Body:
◦ Introduction: Introduces the problem, scope, background context, operational objectives, and
limitations.
◦ Literature Review: Critical analysis of existing published research, establishing the academic gap the
current study addresses.
◦ Research Methodology: Explains the research design, target population, sampling technique,
instruments, data collection procedures, and statistical tests applied.
◦ Data Analysis and Results: Presents findings using tables, charts, and text.
◦ Discussion: Interprets results, connects them back to the literature, discusses implications, and
acknowledges limitations.
◦ Conclusions & Recommendations: Actionable suggestions and insights based on the findings.
3. End Matter:
◦ References / Bibliography: Comprehensive list of all cited sources formatted in standard styles (APA,
IEEE, Harvard).
◦ Appendices: Raw supplemental materials like sample questionnaires, extensive data tables, math
proofs, or code snippets.

21. What is the Concept of Primary Data? Explain sources with examples.
Primary Data is original data gathered for the first time by a researcher to address the specific problem at
hand. It represents unmanipulated, raw empirical evidence collected directly from field subjects.

Primary Sources and Techniques:

1. Questionnaires and Surveys: Highly structured, digital or paper forms featuring closed-ended or open-
ended questions distributed to a large sample pool.

Example: Distributing a Google Form to 500 mobile application users to assess UI/UX satisfaction scores.

2. Interviews: Direct verbal communication between an interviewer and interviewee, ranging from
unstructured dialogues to highly formal, standardized structures.

Example: Conducting one-on-one video calls with 10 Chief Technology Officers (CTOs) to learn about
corporate AI adoption barriers.

MCA 2nd Sem (Pattern 2024) — Research Methodology Reference Guide 9

 3. Observation: The systematic recording of behavioral patterns, events, or processes without actively
communicating with or disrupting the subjects.

Example: A software engineer tracking click-stream maps and user heatmaps on an e-commerce
checkout page to identify drop-off zones.

4. Controlled Experiments: Active scientific setups where independent variables are systematically
modified under controlled conditions to isolate and measure behavioral changes.

Example: Running an A/B test on a landing page design, where Group A sees a blue button and Group B
sees a green button, to evaluate changes in conversion rates.

22. Under what circumstances would you recommend Qualitative and Quantitative
Research? Explain with suitable examples.
The choice between qualitative and quantitative methodologies depends on the nature of the inquiry, the
depth required, and the research objectives.

1. Recommend Qualitative Research When:

• The research objective is to explore subjective experiences, hidden motivations, complex behavioral
dynamics, or cultural phenomena.
• The underlying problem is novel, lacked baseline variables, or requires open-ended discovery instead of
rigid confirmation.
• The focus is on answering "Why?" and "How?" rather than "How many?".

Example: Investigating how software developers experience burnout during sudden transitions to remote
work. This requires open-ended focus groups and deep narrative interviews to capture emotional nuances,
coping strategies, and domestic challenges.

2. Recommend Quantitative Research When:

• The goal is to test specific empirical hypotheses, measure numerical variance, uncover statistical
correlations, or generalize trends across a large population.
• The variables are well-defined, and standardized, objective measurement tools exist.
• The focus is on answering structured, metric questions ("How much?", "How often?", "What percentage?").

Example: Assessing whether implementing a new automated testing tool shortens deployment cycles. The
researcher measures and compares exact deployment times (in hours) and bug counts before and after the
tool's rollout across 100 enterprise software teams.

MCA 2nd Sem (Pattern 2024) — Research Methodology Reference Guide 10

 RESEARCH METHODOLOGY — COMPLETE SOLUTIONS
F.Y. M.C.A. (Management) | Semester - II | Pattern 2024 (RMW 554 MJ)

Q1) a) Differentiate Basic vs. Applied research with example. [5 Marks]

Basic Research (Fundamental/Pure Research): It is driven by a scientist's curiosity or interest in a scientific

question. The main motivation is to expand man's knowledge, not to create or invent something. There is no
obvious commercial value to the discoveries that result from basic research.

Applied Research: It is designed to solve practical problems of the modern world, rather than to acquire
knowledge for knowledge's sake. The goal of the applied scientist is to improve the human condition or solve
specific corporate/organizational problems.

Features Basic Research Applied Research

To expand scientific knowledge and To solve a specific, practical problem or

Purpose
understand fundamental principles. optimize a system.

Theoretical and explanatory; driven by Practical and action-oriented; driven by

Focus
intellectual curiosity. commercial or societal needs.

Universal and broadly applicable across Specific, contextualized, and localized to

Scope
disciplines. a particular problem.

A study looking into how the human Developing an AI-driven chatbot to

brain processes natural language or an automate customer support routing or
Example
investigation into the chemical properties designing a lightweight, durable polymer
of a new polymer compound. shield for mobile devices.

OR

Q1) c) Elaborate Research Process and Steps explain with diagram. [5 Marks]

The research process consists of a series of systematic, sequential steps necessary to effectively carry out research.
Although the steps are presented linearly, research is often an iterative process where steps are revisited.

Step 1: Formulating the Research Problem (Defining Topic & Objectives)

↓
Step 2: Extensive Literature Survey (Reviewing Existing Works)

↓
Step 3: Developing Working Hypotheses (Formulating Testable Statements)

F.Y. M.C.A. (Management) — Research Methodology Answers Page 1 of 12

 ↓
Step 4: Preparing the Research Design (Selecting Methodology & Framework)

↓
Step 5: Determining Sample Design (Selecting Sampling Method & Size)

↓
Step 6: Collecting Data (Primary/Secondary Data Collection Execution)

↓
Step 7: Execution of the Project & Data Analysis (Hypothesis Testing via Stats)

↓
Step 8: Interpretation and Report Writing (Drawing Conclusions & Documentation)

Elaboration of Key Steps:

• Formulation: Stating the core operational issue or theoretical gap clearly.

• Literature Survey: Reading academic papers, journals, and books to build foundation.
• Hypothesis: Proposing tentative answers that can be quantitatively validated.
• Analysis & Testing: Applying mathematical tools (like Chi-Square, t-tests) to accept or reject hypotheses.

OR

Q1) d) How do researchers formulate research questions? Give an example of a well-structured

research question. [5 Marks]

Researchers formulate research questions by moving from a broad topic of interest to a specific, researchable
problem. The standard process involves:

1. Identify a Broad Field: Select an area of active professional or scholarly interest.

2. Conduct Preliminary Literature Review: Identify gaps, contradictions, or unresolved questions in current
published literature.
3. Narrow the Focus: Subdivide the broad area into operational sub-problems based on feasibility, timeline, and
available resources.
4. Apply FINER Criteria: Ensure the question is Feasible, Interesting, Novel, Ethical, and Relevant.

Example of a well-structured research question:

"To what extent does the implementation of automated Deep Learning convolutional neural network (CNN)
models reduce the false-negative diagnosis rates of early-stage breast cancer tumors in radiological
mammograms compared to standard manual double-reading procedures by senior radiologists?"

F.Y. M.C.A. (Management) — Research Methodology Answers Page 2 of 12

 Why it is well-structured: It is highly specific, operationalizes the variables (CNN models vs. manual double-
reading), defines the target outcome (false-negative rates), and explicitly isolates the population domain (early-
stage breast cancer tumors).

F.Y. M.C.A. (Management) — Research Methodology Answers Page 3 of 12

 Q2) a) Compare and contrast experimental and quasi-experimental research designs. [5 Marks]

Both experimental and quasi-experimental research designs aim to identify cause-and-effect relationships by
introducing an intervention or treatment. However, they differ fundamentally on internal validity controls.

Feature Experimental Research Design Quasi-Experimental Research Design

Mandatory. Participants are randomly Absent. Uses pre-existing, intact, or

Random Assignment assigned to either the experimental or naturally occurring groups instead of
control group. random assignment.

May lack a true control group; often

Always features a strict, structurally
Control Group relies on comparison or non-equivalent
identical control group.
groups.

High. Strong ability to isolate the Moderate/Low. Susceptible to selection

Internal Validity independent variable and establish bias and confounding historical/
causality. maturation variables.

Usually conducted in highly controlled Conducted in naturalistic field settings,

Setting Application
laboratory settings. schools, clinics, or social organizations.

OR

Q2) b) Discuss different methods of data collection used in research with examples. [5 Marks]

Data collection methods are broadly categorized into primary data collection (freshly collected original
observations) and secondary data collection (historical or pre-existing records).

• Observation Method: Data is gathered by monitoring and recording human behavioral patterns, events, or
processes systematically without direct interaction. Example: An ethnographer sitting in a hospital waiting
room recording non-verbal patient anxiety indicators.
• Survey / Questionnaire Method: Relying on a structured list of questions distributed to a target sample to
gather attitudes, opinions, or facts. Example: Circulating a Google Form to 500 corporate employees to
measure remote-work burnout metrics.
• Interview Method: Face-to-face, telephonic, or digital synchronous conversational interactions to extract deep
insights. Example: Semi-structured interviews with 10 hospital Chief Medical Officers regarding AI-tool
adoption hurdles.
• Experimental Data Collection: Actively manipulating an independent variable while capturing changing
states of a dependent metric. Example: Recording blood sugar variations before and after administering a new
drug molecule.

OR

F.Y. M.C.A. (Management) — Research Methodology Answers Page 4 of 12

 Q2) c) Define Research and discuss the importance and objectives of research. [5 Marks]

Definition: Research is a structured, systematic, objective, and empirical investigation initiated to discover new
facts, validate existing theories, establish causal frameworks, or find practical resolutions to persistent issues.

Objectives of Research:

1. To Gain Familiarity / Achieve Insights: Exploratory studies to map out ambiguous phenomena.
2. To Portray Characteristics Accurately: Descriptive studies documenting exact attributes of groups or
profiles.
3. To Determine Frequency: Diagnostic studies investigating how often an event interfaces with another.
4. To Test Hypotheses: Causal or experimental testing to validate mathematical correlations.

Importance of Research: It forms the scientific backbone for national socio-economic policy formulations. It
resolves operational market anomalies for commercial industries, catalyzes technological and clinical progress,
and sharpens analytical and critical thinking competencies across academic institutions.

OR

Q2) d) Compare Exploratory design and Descriptive design. [5 Marks]

Dimension Exploratory Design Descriptive Design

To clarify ambiguous problems, To precisely describe traits,

Primary Goal formulate hypotheses, and build baseline demographics, or behaviors of a target
understanding. population.

Highly flexible, unstructured, informal, Rigid, highly structured, formal, and

Structure
and open-ended. planned in advance.

Literature reviews, expert interviews, Structured surveys, observation

Methodology
pilot studies, and small focus groups. checklists, and census databases.

Tentative insights; sets the groundwork Definitive findings; serves as conclusive

Outcome
for conclusive subsequent research. data for corporate decision-making.

F.Y. M.C.A. (Management) — Research Methodology Answers Page 5 of 12

 Q3) a) What is sampling design? Explain its importance of sampling in research. Calculate the
sample size for a population of 5,00,000. Take confidence level as 95% and margin of error as
5%. [P = 0.5] [10 Marks]

Sampling Design: A sampling design is a definite, predefined plan or algorithmic roadmap for obtaining a
representative sample from a given target population before any field data collection begins. It determines the
specific selection logic (e.g., probability vs. non-probability) and calculations governing the sample elements.

Importance of Sampling:

• Cost and Time Efficiency: Consuming data from a smaller subset drastically reduces physical resource
outlays and operational turnaround compared to a full census.
• Higher Accuracy/Depth: Enables researchers to manage quality control, train field staff better, and deploy
highly detailed diagnostic instruments.
• Destructive Testing Capability: In cases where checking involves consuming the element (e.g., testing bullet
velocity or chemical lifespan), sampling is the only viable method.

Sample Size Calculation (Cochran's / Yamane Formulations):

Given parameters:

• Population Size (N) = 500,000

• Confidence Level = 95%, which maps to a standard Z-score (Z) = 1.96
• Margin of Error (e) = 5% = 0.05
• Expected Proportion / Variability (P) = 0.5, meaning Q = 1 - P = 0.5

First, we calculate the sample size for an infinite population (n0) using Cochran's formula:

n0 = (Z² × P × Q) / e²

n0 = (1.96² × 0.5 × 0.5) / 0.05²

n0 = (3.8416 × 0.25) / 0.0025 = 0.9604 / 0.0025 = 384.16

Now, adjust for the finite population (n):

n = n0 / [1 + ((n0 - 1) / N)]

n = 384.16 / [1 + (383.16 / 500,000)]

n = 384.16 / [1 + 0.00076632] = 384.16 / 1.00076632 ≈ 383.86

Rounding up to the nearest whole integer, the required sample size is 384 subjects.

OR

F.Y. M.C.A. (Management) — Research Methodology Answers Page 6 of 12

 Q3) b) Explain Primary Data Collection methods with suitable example. [5 Marks]

Primary data refers to the original information collected firsthand directly from respondents for the specific
research goal at hand. The primary methods include:

1. Questionnaire Method: An instrument consisting of a series of prompt items distributed to respondents.

Example: Delivering a digital 5-point Likert scale survey to assess mobile banking app satisfaction among
senior citizens.
2. Structured/Unstructured Interviews: Direct conversational exchange to uncover deeper cognitive
motivations. Example: Conducting 1-on-1 qualitative interviews with cloud computing experts to determine
emerging cybersecurity threats.
3. Controlled Field Experiments: Manipulating environmental variables to capture direct physical output.
Example: Showing two different landing page variations (A/B testing) to random users and tracking purchase
conversion clicks.

F.Y. M.C.A. (Management) — Research Methodology Answers Page 7 of 12

 Q4) a) Discuss Chi-square test. From the data given below about treatment of 250 patients
suffering from a disease, state whether the new treatment is superior to the conventional
treatment. Formulate null and alternate hypothesis and apply Chi-square test. (Consider: chi-
squared value 0.05 significance level at degree of freedom is = 3.84) [10 Marks]

Discussion: The Chi-square (χ²) test of independence is a non-parametric statistical tool used to evaluate whether
two categorical variables are significantly associated or independent of one another. It compares observed
frequency distributions against expected theoretical frequencies under the assumption of independence.

Step 1: Formulate Hypotheses

• Null Hypothesis (H0): The treatment outcome is independent of the type of treatment. The new treatment is
not superior to the conventional treatment.
• Alternative Hypothesis (H1): The treatment outcome is dependent on the treatment type. The new treatment
is significantly superior to the conventional option.

Step 2: Table of Observed Frequencies (O) and Marginals

Treatment Type Favourable Non-Favourable Total

New Treatment 140 30 170

Conventional 60 20 80

Total 200 50 250

Step 3: Calculate Expected Frequencies (E = (Row Total × Column Total) / Grand Total)

• E(New, Favourable) = (170 × 200) / 250 = 34000 / 250 = 136

• E(New, Non-Favourable) = (170 × 50) / 250 = 8500 / 250 = 34
• E(Conventional, Favourable) = (80 × 200) / 250 = 16000 / 250 = 64
• E(Conventional, Non-Favourable) = (80 × 50) / 250 = 4000 / 250 = 16

Step 4: Compute Chi-square statistic (χ² = ∑ [ (O - E)² / E ])

Cell (Row, Col) Observed (O) Expected (E) (O - E) (O - E)² (O - E)² / E

New, Fav 140 136 4 16 16 / 136 ≈ 0.1176

New, Non-Fav 30 34 -4 16 16 / 34 ≈ 0.4706

Conv, Fav 60 64 -4 16 16 / 64 = 0.2500

Conv, Non-Fav 20 16 4 16 16 / 16 = 1.0000

Summing up the last column: χ² = 0.1176 + 0.4706 + 0.2500 + 1.0000 = 1.8382

F.Y. M.C.A. (Management) — Research Methodology Answers Page 8 of 12

Step 5: Statistical Conclusion

• Calculated Chi-square value: χ² ≈ 1.84

• Critical Table Value at α = 0.05 with df = (2-1)×(2-1) = 1 is given as 3.84.

Since the calculated χ² (1.84) is less than the critical table value (3.84), we fail to reject the null hypothesis
(H0). There is no statistically significant evidence at a 5% significance level to prove that the new treatment is
superior to the conventional treatment.

OR

Q4) b) The population mean of height of 5 year-old boys is 100 cm. A teacher measures height of
25 students, obtaining a mean height of 105 cm & standard deviation 18. Perform the student's
one sample t-test with 5% significance level to calculate whether the true mean is actually
greater than 100cm. (Consider t value at 0.05 significance = 2.064) [10 Marks]

Step 1: Formulate Hypotheses

• Null Hypothesis (H0): μ = 100 cm (The true mean height is equal to 100 cm).

• Alternative Hypothesis (H1): μ > 100 cm (The true mean height is significantly greater than 100 cm — One-
tailed test).

Step 2: Note Down Parameters

• Hypothesized Population Mean (μ) = 100

• Sample Size (n) = 25
• Sample Mean (X̄) = 105
• Sample Standard Deviation (s) = 18
• Degrees of Freedom (df) = n - 1 = 24

Step 3: Calculate the t-statistic Formula

t = (X̄ - μ) / (s / √n)

Substitute values:

t = (105 - 100) / (18 / √25)

t = 5 / (18 / 5) = 5 / 3.6 ≈ 1.3889

Step 4: Interpret the Result

• Calculated t-value = 1.39

• Critical t-value at 5% significance level with df = 24 is 2.064.

F.Y. M.C.A. (Management) — Research Methodology Answers Page 9 of 12

Because the calculated t-statistic (1.39) is less than the critical cutoff parameter (2.064), we fail to reject the null
hypothesis. We conclude that the sample data does not provide enough statistical evidence to claim that the true
mean height of 5-year-old boys is significantly greater than 100 cm.

F.Y. M.C.A. (Management) — Research Methodology Answers Page 10 of 12

 Q5) a) Suggest an appropriate framework for organizing a research report on "Machine
learning applications in healthcare". [10 Marks]

An ideal architectural blueprint for an advanced post-graduate engineering or management research report must
comply with formal scientific disclosure layout requirements as structured below:

1. Preliminary Pages
• Title Page: Explicitly states topic, researcher details, institution, and submission year.
• Certificate of Authenticity: Validated clearance endorsement by academic guide.
• Abstract: A compact 250-word synthesis of problem background, methodologies, ML models deployed,
core test metrics (accuracy/AUC), and clinical implications.
• Table of Contents & List of Tables/Figures: Operational page indexing.

2. Main Body Contents

• Chapter 1: Introduction: Background information on medical imaging, diagnostics, electronic health
records (EHR); objectives definition, and statement of intent.
• Chapter 2: Literature Review: Exhaustive evaluation of current clinical models, CNNs, Transformers, or
random forest applications; isolating existing gaps in model interpretability or data imbalances.
• Chapter 3: Research Methodology & Engineering Pipeline: Details of data source (e.g., MIMIC-III,
open Kaggle clinical sets), preprocessing transformations, feature extraction, architectural hyperparameter
selections, and validation splits.
• Chapter 4: Results and Model Analysis: Disclosing training graphs, confusion matrices, ROC curves,
execution time profiles, and standard error rates.
• Chapter 5: Discussion: Critical interpretation of computational achievements against medical
benchmarks, model limitations (bias, black-box issues), and ethical data privacy vectors.
• Chapter 6: Conclusion & Future Scope: Definitive final output declarations, regulatory FDA/clinical
challenges, and algorithmic pathways forward.

3. Postliminary Sections
• References / Bibliography: APA or IEEE indexed sourcing list of academic artifacts.
• Appendices: Source code repositories, specific hardware configurations, or extended statistical calculation
tables.

OR

Q5) b) Prepare an outline for a research on topic "Role of AI in Healthcare". [10 Marks]

Below is a structured blueprint layout outlining a dedicated thesis project targeting artificial intelligence
deployment vectors within healthcare systems.

F.Y. M.C.A. (Management) — Research Methodology Answers Page 11 of 12

 1. Title: "Assessing the Impact of Deep Learning Architectures on Diagnostic Accuracy and Patient Triage
Efficiency in Modern Clinical Ecosystems"

2. Objectives:

• To evaluate the performance optimization delta of Convolutional Neural Networks (CNNs) in identifying
pulmonary anomalies from chest X-rays.
• To analyze the reduction in administrative operational triage response time for emergency critical care
admissions.

3. Hypotheses:

• Null Hypothesis (H0): AI-driven triage implementation produces no significant change in emergency
department waiting times.
• Alternative Hypothesis (H1): AI-driven triage systems significantly lower patient processing times
compared to traditional nurse-led triage methods.

4. Research Gap:
While existing research extensively optimizes abstract laboratory algorithm accuracy, there remains a critical
operational research gap regarding how real-world clinical data drift and systemic clinician distrust alter
model deployment efficacy in high-pressure emergency clinical units.

5. Data Collection:
Retrospective acquisition of anonymized historical radiological image streams alongside active, time-tracked
log tracking entries across a multi-center public hospital deployment trial.

6. Sampling Design:
Stratified random sampling across distinct clinical demographic units (categorized by age blocks and acute
admission severities), evaluating a target sample size of n = 500 unique patient diagnostic records.

7. Results & Discussion:

Exposing achieved validation accuracy benchmarks (e.g., 94.2% accuracy), comparative run-time bar graphs,
and addressing computational limitations such as adversarial noise vulnerability and the ethical demand for
Explainable AI (XAI) overlays.

8. Conclusion:
AI systems function efficiently as powerful collaborative diagnostic amplifiers rather than standalone
replacements for medical experts. Regulatory framework standards must evolve to address liability dynamics
before large-scale hospital integration.

F.Y. M.C.A. (Management) — Research Methodology Answers Page 12 of 12

 COMPREHENSIVE SOLUTION MANUAL
F.Y. M.C.A. (Management) Semester - II
RMW 554 MJ - RM21 : RESEARCH METHODOLOGY

Course: First Year Master of Computer Applications Pattern: 2024 Pattern

(Management)

Subject Code: [6432]-2012 / PD2857 Max. Marks: 50 Marks

Duration: 2½ Hours Academic Level: Post Graduate

Graduate Level

Instructions to the Candidates:

1. All questions are fully answered in this text compiled for reference study.
2. In the actual exam layout, options (OR questions) are provided. Both paths are thoroughly resolved here.
3. Calculations are detailed step-by-step to demonstrate standard formulation.

F.Y. M.C.A. (Management) - Research Methodology Solution Book Page 1 of 17

 Q1) a) Describe research process with suitable diagram in detail. [5]

The research process consists of a series of systematic, sequential steps necessary to effectively carry out research.
While the steps are presented linearly, the actual process is iterative, often requiring researchers to revisit previous steps
based on new insights.

1. Formulating the Research Problem

2. Extensive Literature Review

3. Development of Working Hypotheses

4. Preparing the Research Design

5. Determining Sample Design

6. Data Collection (Execution of Project)

7. Analysis of Data & Hypothesis Testing

8. Generalization, Interpretation & Reporting

Detailed Explanation of the Steps:

• 1. Formulating the Research Problem: The first step is to identify and clearly define a broad statement of a
general question, then narrow it down into a specific research problem. This involves conceptualizing the topic,
understanding its relevance, and stating operational objectives.
• 2. Extensive Literature Review: Researchers review existing conceptual framework knowledge, books, academic
journals, and past empirical reports. This establishes context, prevents duplication, and highlights critical research
gaps.
• 3. Formulation of Working Hypotheses: A hypothesis is a tentative predictive proposition tentative to empirical
verification. It guides the researcher's focus towards factual data collection and appropriate testing measures.

F.Y. M.C.A. (Management) - Research Methodology Solution Book Page 2 of 17

 • 4. Preparing the Research Design: This is a conceptual structure or blueprint within which research is conducted.
It determines whether the research is exploratory, descriptive, diagnostic, or experimental, specifying resource
allocation, timing, and logical sequencing.
• 5. Determining Sample Design: Since testing an entire population is rarely feasible, a precise sampling plan
(probability or non-probability) must be selected to extract a representative subset.
• 6. Data Collection & Execution: The plan is executed via fieldwork to gather data using instruments like
questionnaires, interviews, observations, or database queries.
• 7. Data Analysis & Hypothesis Testing: Raw data is edited, coded, classified, and tabulated. Statistical tests (e.g.,
Chi-square, t-test, ANOVA) are applied to determine if the findings support or reject the stated hypotheses.
• 8. Interpretation & Reporting: The final stage converts empirical outcomes into theoretical insights,
contextualizes findings, and formally presents them in a structured research report.

Q1) b) Explain importance of Literature review in research. [5]

A literature review is an intensive, evaluative account of existing published works by scholars and researchers on a
specific topic. Its paramount importance in scientific inquiry is summarized through the following dimensions:

• Identification of Research Gaps: By mapping out existing knowledge, it uncovers variables that have been
ignored, methodological weaknesses in past studies, or unexamined demographics, providing the justification for a
new project.
• Avoiding Duplication: It prevents researchers from wasting resources to rediscover facts that are already robustly
established in the academic domain.
• Methodological Guidance: It reveals the analytical frameworks, research designs, sampling approaches, and data
collection tools utilized by earlier researchers, providing a roadmap for one's own research methodology.
• Theoretical Grounding: It provides a deep conceptual understanding of the theories, variables, and paradigms
governing the area of study, helping to build strong hypotheses.
• Contextualizing Findings: Once data collection and analysis are complete, the literature review acts as a
benchmark, enabling the researcher to compare their results with established literature to point out similarities or
deviations.

OR

Q1) c) Differentiate qualitative and quantitative research methods with suitable example. [5]

The core differences between qualitative and quantitative paradigms are outlined systematically in the table below:

F.Y. M.C.A. (Management) - Research Methodology Solution Book Page 3 of 17

 Feature Qualitative Research Quantitative Research

To understand underlying To quantify data and generalize

reasons, motivations, meanings, findings across a population.
Focus / Objective
and experiences. Explores the Measures the "What", "How
"Why" and "How". much", and "How many".

Non-numerical text, descriptive Numerical metrics, structured

Data Type words, video transcripts, images, measurements, statistics, counts,
or observations. and percentages.

Inductive (bottom-up Deductive (top-down verification

Approach
development of theory). of hypothesis/theory).

Small, purposeful, and Large, representative samples to

Sample Size
information-rich selections. ensure statistical validity.

Thematic analysis, content Statistical processing,

Data Analysis synthesis, or interpretative mathematical modelling, and
summarization. descriptive/inferential analysis.

Illustrative Examples:

• Qualitative Example: Conducting in-depth focus group interviews with ten users to explore their subjective
emotional experiences, frustrations, and cognitive difficulties while using a new enterprise software application.
• Quantitative Example: Distributing a structured 5-point Likert scale questionnaire to 500 users to calculate the
mean usability rating, system loading speeds, and compiling the exact percentage distribution of operational errors.

Q1) d) What is Research Methodology? Explain objectives of research. [5]

Research Methodology: Research methodology refers to the systematic, theoretical blueprint used to solve a research
problem. It is the overall strategy that shapes the choice and application of specific methods. It goes beyond the methods
themselves to encompass the rationale and philosophical principles behind why a researcher chooses particular
techniques, instruments, and statistical treatments, detailing how data is extracted, processed, and validated.

Objectives of Research:

The primary purpose of research is to discover answers to inquiries through systematic procedures. The core objectives
can be categorized as follows:

• To Gain Familiarity or Achieve New Insights (Exploratory/Formulative Studies): Geared towards investigating
an under-explored field to generate preliminary insights and clarify complex concepts for deeper research.
• To Portray Characteristics Accurately (Descriptive Studies): Aimed at precisely mapping the profiles,
behaviors, or situational characteristics of an individual, group, or system.

F.Y. M.C.A. (Management) - Research Methodology Solution Book Page 4 of 17

 • To Determine Frequency of Occurrence (Diagnostic Studies): Concerned with identifying how often a specific
event occurs or assessing its operational association with other variables.
• To Test Causal Relationships (Hypothesis-Testing/Experimental Studies): Set up to rigorously examine a
dependency relationship between independent variables and dependent variables under controlled conditions.

F.Y. M.C.A. (Management) - Research Methodology Solution Book Page 5 of 17

 Q2) a) Elaborate types of research with example. [10]

Research paradigms can be classified based on their underlying objectives, context, application, and structural
procedures. Below are the primary types explained with corresponding examples:

1. Descriptive vs. Analytical Research:

◦ Descriptive: Focuses on describing the current state of affairs as it exists, without control over variables (Ex-
post facto).
Example: A market survey determining the current percentage of people using digital wallets in a particular
city.
◦ Analytical: Uses facts or data already available to critically analyze and evaluate a phenomenon.
Example: Evaluating why digital wallet usage spiked during a specific fiscal quarter by analyzing economic
indicators.

2. Applied vs. Fundamental (Pure/Basic) Research:

◦ Applied: Aims at finding an immediate solution to a concrete operational problem facing an organization,
industry, or society.
Example: Designing a specific patch to resolve a critical security vulnerability in an e-commerce platform.
◦ Fundamental: Primarily concerned with generalizations and formulation of a theory, seeking to broaden the
boundaries of knowledge.
Example: Developing a new abstract mathematical theorem or a novel theoretical framework for quantum
computing.

3. Quantitative vs. Qualitative Research:

◦ Quantitative: Based on measuring quantity or amount, applicable to phenomena that can be expressed
numerically.
Example: Tracking the annual revenue growth rate across 100 IT firms to identify industry trends.
◦ Qualitative: Concerned with qualitative phenomena, focusing on subjective assessments of behavior, attitude,
or opinions.
Example: An ethnographic study analyzing how corporate culture impacts employee morale through deep
behavioral observations.

4. Conceptual vs. Empirical Research:

◦ Conceptual: Related to abstract ideas or theories. Used by philosophers or thinkers to develop new concepts or
reinterpret existing ones.
Example: Reinterpreting classic human motivation theories within the modern context of remote work
frameworks.
◦ Empirical: Relies on experiential or direct observations and experiments. It is data-based, coming up with
conclusions capable of verification by observation or experiment.
Example: Testing a new clinical drug on a group of patients and measuring recovery rates against a control
group.

F.Y. M.C.A. (Management) - Research Methodology Solution Book Page 6 of 17

 OR

Q2) b) What is hypothesis? Prepare a hypothesis (null & alternate) for following situations.
i) Researcher wants to understand the impact of social media usage on children's behaviour.
ii) Researcher wants to understand the lectures attended by first-year students and its effect on
final exam scores. [5]

Hypothesis: A hypothesis is a formal, tentative statement that proposes a predictive relationship between two or more
variables. It acts as an operational proposition that can be empirically tested using statistical tools to confirm or refute
its validity.

Formulation for Specific Situations:

Situation i): Impact of social media usage on children's behaviour.

• Null Hypothesis (H0): There is no significant impact of social media usage on children's behavior. (i.e., varying
levels of social media usage do not alter behavioral metrics).
• Alternative Hypothesis (H1): There is a significant impact of social media usage on children's behavior. (i.e.,
higher social media usage is correlated with observable changes in behavior).

Situation ii): Lectures attended by first-year students and its effect on final exam scores.

• Null Hypothesis (H0): There is no significant relationship between the number of lectures attended by first-year
students and their final exam scores. (i.e., attendance does not affect performance).
• Alternative Hypothesis (H1): There is a significant positive relationship between the number of lectures attended
by first-year students and their final exam scores. (i.e., higher attendance leads to an increase in exam scores).

Q2) c) What are the key components of a well-structured research design? [5]

A well-structured research design serves as the operational framework of the study. Its key components include:

• 1. Title and Problem Statement: A concise, unambiguous description of the core research problem under
investigation and the scope of the inquiry.
• 2. Purpose and Objectives of the Study: A clear statement outlining what the study aims to achieve, its
importance, and the specific research questions or hypotheses it addresses.
• 3. Population and Sampling Design: Defining the target population, sampling frame, sample size calculation, and
the exact sampling method (probability vs. non-probability) to ensure proper representation.
• 4. Operationalization of Variables: Explicitly identifying independent, dependent, confounding, and intervening
variables, and defining how they will be measured or manipulated.
• 5. Data Collection Strategy: Specifying the sources of data (primary or secondary) and the methods used to collect
it (e.g., surveys, experiments, interviews, observations).

F.Y. M.C.A. (Management) - Research Methodology Solution Book Page 7 of 17

 • 6. Analysis and Statistical Tools: Outlining the methods for editing, coding, and tabulating data, alongside the
statistical tests (e.g., t-test, Chi-square, regression) used to evaluate hypotheses.
• 7. Logistics and Resource Management: A clear timeline (such as a Gantt chart) detailing key phases, alongside
budget constraints and resource allocations.

F.Y. M.C.A. (Management) - Research Methodology Solution Book Page 8 of 17

 Q3) a) Explain Probability and Non-probability Sampling methods with suitable examples. [10]

Sampling is the process of selecting a subset of individuals from a population to estimate characteristics of the whole
group. Sampling techniques are divided into probability and non-probability methods.

1. Probability Sampling Methods:

Every element in the target population has a known, non-zero chance of selection. This minimizes selection bias and
supports statistical generalization.

• Simple Random Sampling: Every member has an equal probability of being chosen. Selection is usually done via
lottery or random number generators.
Example: Pulling 50 student IDs out of a hat containing all 1,000 enrolled college students.
• Stratified Random Sampling: The population is divided into mutually exclusive subgroups (strata) based on
shared traits (e.g., gender, age, income), and random samples are drawn from each stratum.
Example: Dividing a university population into Freshmen, Sophomores, Juniors, and Seniors, then randomly
selecting 50 students from each group to assess satisfaction levels.
• Systematic Sampling: Elements are selected from an ordered sampling frame at regular, predetermined intervals
determined by a sampling interval k = N / n.
Example: Selecting every 10th customer who walks into a retail store, starting from a randomly selected initial
customer.
• Cluster Sampling: The population is divided into clusters (often geographical), and entire clusters are randomly
selected for full analysis.
Example: Selecting 5 city blocks at random from an urban layout and surveying every household within those
selected blocks.

2. Non-Probability Sampling Methods:

Selection relies on the subjective judgment of the researcher rather than random chance. These methods are typically
used for exploratory studies where statistical generalization is not the primary goal.

• Convenience Sampling: Data is collected from population members who are most accessible to the researcher.
Example: A researcher standing outside a shopping mall and interviewing anyone who passes by.
• Purposive / Judgmental Sampling: The researcher intentionally selects specific individuals based on their
expertise or characteristics relevant to the study.
Example: Interviewing only senior IT managers with over 15 years of experience to study cloud infrastructure
trends.
• Quota Sampling: The population is segmented into subgroups, and a non-random quota of participants is chosen
from each group until the target numbers are met.
Example: Instructing field workers to collect data from 40 males and 60 females on the street, without randomized
selection.

F.Y. M.C.A. (Management) - Research Methodology Solution Book Page 9 of 17

 • Snowball Sampling: Initial participants help recruit additional eligible subjects from their personal or professional
networks. This is especially useful for reaching hard-to-access populations.
Example: Conducting research on rare medical disorders where existing patients introduce the researcher to other
diagnosed individuals.

OR

Q3) b) Suppose you are asked to select a sample for is conducting a survey on students'
learning habits. Which sampling method would be most suitable and why? What will be sample
size for the same? [10]

1. Recommended Sampling Method: Stratified Random Sampling

To conduct a rigorous, objective survey on students' learning habits across an institution or university, Stratified
Random Sampling is the most suitable method.

Rationale (Why?):

• Heterogeneity of Learning Habits: Learning habits vary significantly across different academic disciplines (e.g.,
Engineering vs. Management vs. Fine Arts) and across distinct year levels (e.g., First-year undergraduate vs. Final-
year Postgraduate).
• Proportional Representation: Dividing the overall student population into homogeneous strata based on their
specific Department/Stream or Academic Year ensures that smaller departments are not underrepresented, which
can happen with simple random sampling.
• Enhanced Statistical Precision: Minimizing variances within each stratum leads to a more precise estimation of
overall student learning behaviors.

2. Determining the Sample Size:

To calculate a mathematically valid sample size (n) without knowing the exact total population, we use Cochran's
Sample Size Formula for an infinite or large population:

n = (Z2 × p × q) / e2

Where parameters are defined by standard convention:

• Z = Standard normal deviation value corresponding to the desired confidence level. For a standard 95% confidence
level, Z = 1.96.
• p = Estimated proportion of an attribute present in the population. Lacking previous empirical benchmarks, we
assume maximum variability: p = 0.5.
• q = 1 - p = 1 - 0.5 = 0.5.
• e = Acceptable margin of error (precision level). We set a standard margin of error: e = 0.05 (i.e., ±5%).

F.Y. M.C.A. (Management) - Research Methodology Solution Book Page 10 of 17

Step-by-step Calculation:

n = (1.962 × 0.5 × 0.5) / (0.05)2

n = (3.8416 × 0.25) / 0.0025
n = 0.9604 / 0.0025 = 384.16

Rounding up to the nearest whole integer yields an ideal sample size of 385 students. If the university's total population
(N) is small (e.g., N = 2,000), this can be adjusted using Cochran's finite population correction formula:

nadjusted = n / [1 + ((n - 1) / N)]

F.Y. M.C.A. (Management) - Research Methodology Solution Book Page 11 of 17

 Q4) a) Discuss Chi-square test. From the following data, find out whether there is any
relationship between Gender (Male/Female) & Colour Preferences:

Colour Males Females Total

Pink 10 40 50

Black 70 30 100

Yellow 30 20 50

Total 110 90 200

(consider: chi-squared value 0.05 significance level at degree of freedom 2 is = 5.99) [10]

Discussion of Chi-Square (χ2) Test:

The Chi-square test of independence is a non-parametric statistical tool used to determine if there is a significant
association between two categorical variables. It compares observed frequencies (O) from empirical data against
expected frequencies (E) calculated under the assumption that the variables are independent. The formula is:

χ2 = Σ [ (O - E)2 / E ]

Step 1: Formulate Hypotheses

• Null Hypothesis (H0): Gender and color preferences are independent of each other (there is no significant
relationship between them).
• Alternative Hypothesis (H1): Gender and color preferences are dependent (there is a significant relationship
between them).

Step 2: Calculate Expected Frequencies (E)

The expected frequency for any cell is calculated as: E = (Row Total × Column Total) / Grand Total. Given the Grand
Total is 200, Male Total is 110, and Female Total is 90:

• Pink-Male: (50 × 110) / 200 = 27.5; Pink-Female: (50 × 90) / 200 = 22.5
• Black-Male: (100 × 110) / 200 = 55.0; Black-Female: (100 × 90) / 200 = 45.0
• Yellow-Male: (50 × 110) / 200 = 27.5; Yellow-Female: (50 × 90) / 200 = 22.5

Step 3: Compute Chi-Square Value (χ2)

Cell (Color-Gender) Observed (O) Expected (E) (O - E) (O - E)² (O - E)² / E

Pink - Male 10 27.5 -17.5 306.25 11.136

Pink - Female 40 22.5 17.5 306.25 13.611

F.Y. M.C.A. (Management) - Research Methodology Solution Book Page 12 of 17

 Cell (Color-Gender) Observed (O) Expected (E) (O - E) (O - E)² (O - E)² / E

Black - Male 70 55.0 15.0 225.00 4.091

Black - Female 30 45.0 -15.0 225.00 5.000

Yellow - Male 30 27.5 2.5 6.25 0.227

Yellow - Female 20 22.5 -2.5 6.25 0.278

Summing up the final column:

χ2 = 11.136 + 13.611 + 4.091 + 5.000 + 0.227 + 0.278 = 34.343

Step 4: Decision and Interpretation

• Calculated χ2 value = 34.343

• Critical Table value at df = 2 and α = 0.05 is 5.99

Since the calculated Chi-square value (34.343) is much greater than the critical table value (5.99), we reject the null
hypothesis (H0) and accept the alternative hypothesis (H1).

Conclusion: There is a highly significant relationship between gender and color preferences.

OR

Q4) b) What is t-test, When it is used? Determine whether the average weight of sample of 20
mangoes is significantly different from population average weight of 70 gms. The sample mean
weight is 70.55gms & sample standard deviation is 2.82 gm. Apply one sample t test for the
above example (consider t value at 0.05 significance = 2.093). [10]

Concept of t-test and Conditions for Use:

A t-test is a parametric statistical test used to compare means and determine if they are significantly different from one
another. It is used when:

• The sample size is small (n < 30).

• The population standard deviation (σ) is unknown.
• The underlying population distribution is assumed to be normal.

Mathematical Application: One-Sample t-test

F.Y. M.C.A. (Management) - Research Methodology Solution Book Page 13 of 17

Step 1: Formulate Hypotheses

• Null Hypothesis (H0): μ = 70 gms (The average weight of mangoes is not significantly different from the
population mean).
• Alternative Hypothesis (H1): μ ≠ 70 gms (The average weight of mangoes is significantly different from the
population mean).

Step 2: Note Given Variables

• Population Mean (μ) = 70 gms

• Sample Size (n) = 20
• Sample Mean (X̄) = 70.55 gms
• Sample Standard Deviation (s) = 2.82 gms
• Degrees of Freedom (df) = n - 1 = 20 - 1 = 19

Step 3: Compute the t-statistic

The formula for a one-sample t-test is:

t = (X̄ - μ) / (s / √n)

First, calculate the standard error of the mean:

SE = s / √n = 2.82 / √20 = 2.82 / 4.472 = 0.6306

Now, compute the t-value:

t = (70.55 - 70) / 0.6306 = 0.55 / 0.6306 = 0.872

Step 4: Decision and Interpretation

• Calculated absolute |t| value = 0.872

• Critical Table t value at df = 19 and α = 0.05 is 2.093

Since the calculated t-value (0.872) is less than the critical table value (2.093), we fail to reject the null hypothesis (H0).

Conclusion: The average weight of the sample of mangoes (70.55 gms) is not significantly different from the
population average weight of 70 gms. The observed difference is statistically negligible and can be attributed to random
sampling variation.

F.Y. M.C.A. (Management) - Research Methodology Solution Book Page 14 of 17

 Q5) a) Select a research topic within a domain of your choice (e.g., Artificial Intelligence,
Machine Learning, Cloud Computing, Internet of Things (IOT), Block Chain, Social Media,
Cybersecurity, etc. or any topic). Prepare a research outline which includes a Title, abstract,
introduction, literature review, Objectives, Hypothesis, methodology, results & discussion,
conclusion, and references. [10]

Research Outline Portfolio

1. Title: Optimizing Edge-Computing Resource Allocation for Real-Time Traffic Management in Smart Cities Using
Machine Learning Frameworks.

2. Abstract:
Modern smart city infrastructures produce large volumes of geospatial data that require immediate processing.
Traditional cloud architectures introduce latency bottlenecks that limit real-time traffic control. This study proposes an
adaptive machine learning resource allocation framework designed for edge nodes. By dynamically distributing
computing tasks based on localized traffic predictions, the framework minimizes system latency. Simulation results
indicate a 24% reduction in data processing latency compared to centralized cloud models, validating the feasibility of
edge-layer intelligence for smart infrastructure management.

3. Introduction:
Urban traffic congestion is a major source of economic inefficiency and environmental emissions. While modern IoT
sensors collect real-time traffic metrics, transferring this data to centralized servers leads to processing delays.
Integrating Machine Learning (ML) at the network edge allows for localized processing and immediate traffic
adjustments. This study outlines a resource allocation framework tailored to the computing constraints of edge-
computing nodes.

4. Literature Review:
Recent studies by Smith et al. (2022) highlight that centralized cloud environments struggle with network congestion
during peak commuting hours. Kumar (2023) demonstrated that edge computing alleviates bandwidth constraints by
processing data near the source. However, dynamic workload balancing across heterogeneous edge nodes remains an
open technical challenge, which this research aims to address.

5. Objectives:

• To design an automated, low-overhead ML model capable of forecasting short-term computing workloads at

localized edge nodes.
• To build a dynamic resource allocation algorithm that minimizes processing latency for automated traffic control
systems.

6. Hypothesis:

• Null Hypothesis (H0): The proposed ML-driven edge allocation model shows no significant reduction in
processing latency compared to standard round-robin cloud methods.

F.Y. M.C.A. (Management) - Research Methodology Solution Book Page 15 of 17

 • Alternative Hypothesis (H1): The proposed ML-driven edge allocation model significantly reduces processing
latency compared to standard round-robin cloud methods.

7. Methodology:
This study utilizes an experimental simulation design. An urban traffic environment comprising 50 intersection nodes is
emulated. We implement a lightweight Random Forest regression algorithm on simulated edge processors to forecast
incoming data traffic. Performance is benchmarked against centralized cloud models and non-predictive edge allocation
strategies over 100 simulation hours.

8. Expected Results & Discussion:

The predictive model is expected to achieve a mean absolute percentage error (MAPE) below 6% for localized traffic
forecasting. Results demonstrate that the dynamic model balances computing workloads efficiently, keeping average
latency under 15ms, well below the 65ms baseline of centralized architectures. This confirms the value of edge-layer
intelligence during peak usage periods.

9. Conclusion:
Deploying machine learning models for resource allocation at the edge significantly improves the responsiveness of
smart city systems. Transitioning from centralized architectures to localized edge processing provides the low-latency
framework necessary for automated traffic management.

10. References:

1. Kumar, A. (2023). Edge Computing Paradigms in Modern IoT Infrastructures. Academic Press.
2. Smith, J., & Jones, M. (2022). Cloud Latency Obstacles in Real-Time Urban Systems. Journal of Smart Cities Research, 14(2),
112-128.

OR

Q5) b) Prepare an outline for a research on topic "Impact of Social Media on Indian Youth"
with proper title, objectives, hypothesis, Research Gap, Data Collection, sampling, Result and
Discussion and Conclusion. [10]

Research Outline Portfolio

1. Title: Digital Consumption Paradigms: Assessing the Impact of Short-Form Video Social Media Platforms on the
Academic Attention Span of Indian Youth.

2. Objectives:

• To measure the average daily hours spent by Indian undergraduate students on short-form video streaming
platforms (e.g., Instagram Reels, YouTube Shorts).
• To evaluate the statistical relationship between daily social media usage and students' self-reported academic focus
and concentration levels.

F.Y. M.C.A. (Management) - Research Methodology Solution Book Page 16 of 17

3. Hypothesis:

• Null Hypothesis (H0): There is no significant relationship between daily time spent on short-form social media
videos and the academic attention span of Indian youth.
• Alternative Hypothesis (H1): There is a significant negative relationship between daily time spent on short-form
social media videos and the academic attention span of Indian youth.

4. Research Gap:
While existing literature extensively explores the broader psychological impacts of social media on mental health, there
is a lack of research specifically focusing on how short-form vertical videos (under 60 seconds) affect cognitive
retention and academic attention spans within the context of the Indian higher education system.

5. Sampling Design:
The target population consists of undergraduate students aged 18 to 23 across major colleges in an urban hub. This
study utilizes Stratified Random Sampling to select participants. The population is stratified by stream (Arts, Science,
Commerce), drawing a proportional random sample from each to achieve a total sample size of 400 respondents.

6. Data Collection Methodology:

Primary data is gathered via a structured online questionnaire distributed through institutional portals. The instrument
includes automated time-tracking metrics (self-reported via device screen-time screenshots) alongside the standardized,
5-point Likert scale Adult Attention Deficit Assessment index to measure cognitive attention span.

7. Expected Result and Discussion:

Descriptive statistics are expected to show an average daily consumption of 3.2 hours of short-form video content
among participants. Correlation analysis is anticipated to reveal a significant negative correlation coefficient (r = -0.64,
p < 0.01) between daily usage and sustained attention spans, supporting the alternative hypothesis. The discussion
emphasizes that rapid visual pacing may contribute to reduced tolerance for traditional lecture formats.

8. Conclusion:
High consumption of short-form social media content is associated with a measurable decrease in academic attention
span among urban Indian youth. These findings suggest a need for educational institutions to update pedagogical
methods and promote digital wellness initiatives to address changing cognitive habits.

F.Y. M.C.A. (Management) - Research Methodology Solution Book Page 17 of 17