INTERNAL ASSIGNMENT

NAME SOURABH KUMAR

ROLL NO 251410505050

SESSION JULY-AUGUST 2025

PROGRAM MASTER OF BUSINESS
ADMINISTRATION (MBA)
SEMESTER II
COURSE CODE & NAME DMBA214 BUSINESS RESEARCH
METHODS (R/PYTHON)
 SET I

Q.1)

Ans- A research design provides the structural outline that shapes how an investigation will
be approached, ensuring consistency between objectives, data requirements, and analytical
methods. In business contexts, this framework supports managers and researchers in
generating insights that are systematic, evidence-based, and relevant to organizational
challenges. A suitable design strengthens the reliability of results and enhances the value of
decisions derived from research findings.

Exploratory Research Design

Exploratory design is adopted when the problem is unstructured or when preliminary insight
is needed before establishing formal hypotheses. This approach makes use of open
discussions, expert consultations, projective techniques, and review of existing information.
A company might apply this design during the early stages of concept development for a new
service, using informal conversations with customers to uncover unmet needs. Consultancy
firms also use exploratory research to map emerging trends before advising clients on
strategic directions.

Descriptive Research Design

Descriptive design generates detailed representations of a situation by measuring and
documenting variables as they exist. Data are gathered through structured surveys,
observation, or systematic documentation of behaviors and conditions. In business
applications, a firm may utilize descriptive research to analyze market share distribution
within an industry or to compare service satisfaction across branches. A banking institution
may rely on descriptive studies to understand the usage patterns of its digital platforms among
different age groups.

Causal (Experimental) Research Design

Causal design aims to establish relationships between variables by observing how changes in
one factor produce variations in another. This method is widely employed in business
experimentation, including A/B testing, controlled trials, and prototype testing. A technology
company may alter interface features for selected users to assess the impact on engagement.
A beverage manufacturer may experiment with two packaging styles in separate regions to
determine their influence on sales performance.

Diagnostic Research Design

Diagnostic design supports studies where identifying the underlying cause of a business
problem is essential. It progresses through recognizing the issue, investigating its drivers, and
reviewing possible solutions. Organizations use this design when confronting challenges such
as declining productivity or rising customer complaints. For example, a hotel chain may
analyze guest feedback trends to identify whether dissatisfaction stems from service delays,
staff behavior, or facility issues.

Cross-Sectional and Longitudinal Designs

Cross-sectional design collects information from a population at a single moment, suitable for
assessing current attitudes or market conditions. Longitudinal design records data repeatedly
over time, offering insight into patterns of change. A retail brand may run a cross-sectional
study to understand festival shopping intentions, while a longitudinal study may track loyalty
program effectiveness across several months.
Q.2)

Ans- RStudio functions as an integrated environment that organizes the entire workflow of
programming and data analysis in R. It enriches base R by offering a cleaner interface,
structured layout, and tools that simplify both coding and interpretation of results. The
platform includes capabilities such as syntax-aware editing, customizable themes, code
snippets, and interactive visualization windows. These elements reduce common coding
errors and support efficient development of statistical models, data manipulation tasks, and
research documentation. RStudio also enables the creation of reproducible analyses through
its support for R Markdown, Quarto, Shiny applications, and well-structured project folders.

Its built-in package manager assists users in installing, updating, and loading libraries without
relying solely on command-line methods. The workspace organization, version control
integration, and debugging utilities further enhance the analytical experience, allowing
researchers to track progress, refine scripts, and maintain consistent documentation. These
combined features extend the power of base R and offer a more intuitive approach to complex
research activities.

Source Pane
The Source pane provides a dedicated space to write and edit scripts, functions, markdown
documents, and analytical reports. It helps users separate exploratory commands from
finalized code by allowing files to be saved and reused across sessions. Features such as
syntax highlighting, line numbering, comment toggling, and execution of selected code
sections make the process more structured. It supports detailed workflows, making it easier
for business researchers to document steps, refine code, and generate organized outputs.

Console Pane
The Console pane displays the real-time interaction with the R interpreter. Users can type
commands directly and observe immediate output. This area is essential for testing ideas,
running quick calculations, or checking function behavior before incorporating the code into a
formal script. The console reflects the core functionality of base R while integrating smoothly
with the rest of the interface, allowing results to appear in relevant panes without additional
configuration.

Environment/History Pane
The Environment tab presents an overview of all active objects created during the session,
including data frames, variables, and imported files. It assists users in monitoring memory
usage, reviewing object structures, and clearing unnecessary items. The History tab records
previously executed commands, offering a valuable reference for reconstructing analytical
steps or identifying patterns in the workflow. These features support transparency and
methodical research practices.

Files/Plots/Packages/Help Pane
This multipurpose pane supports essential tasks such as browsing project directories, viewing
generated plots, managing installed libraries, and accessing documentation. The Files tab
facilitates dataset handling, the Plots tab displays visual summaries, the Packages tab assists
with library management, and the Help tab provides explanations of functions and syntax.
These resources ensure that analytical work remains organized and well-supported.
Q.3)

Ans- Data collection methods differ in the extent of control exercised during the information-
gathering process. Structured, semi-structured, and unstructured methods represent three
distinct approaches, each offering a different balance of consistency, depth, and flexibility. In
business research, these approaches influence the type of data obtained and the kinds of
insights that organizations can derive for strategic or operational decisions.

Structured Data Collection Methods

Structured methods rely on fixed formats where all respondents answer the same set of
questions in an identical sequence. The responses are usually numerical or categorical,
allowing for straightforward comparison and statistical processing. Examples include closed-
ended questionnaires, rating scales, checklists, and standardized observation forms.
Businesses use structured methods when precision and uniformity are essential. A company
evaluating customer satisfaction across multiple branches may employ a fixed survey to
compare results reliably. These methods support large datasets and objective measurement,
making them suitable for market assessments, employee feedback summaries, or consumer
preference studies. However, the opportunity for respondents to provide deeper explanations
is limited due to the rigid response structure.

Semi-Structured Data Collection Methods

Semi-structured methods blend predetermined questions with the flexibility to explore topics
that arise during the interaction. The researcher follows a broad interview guide but has the
freedom to modify the sequence or add probing questions. Examples include semi-structured
interviews, partially guided discussions, and flexible observation protocols.
In business settings, these methods are used when organizations require both structured
insights and contextual understanding. A firm investigating service quality issues may use
semi-structured interviews to uncover experiences that standardized surveys cannot fully
capture. This approach helps identify patterns while still allowing respondents to provide
detailed narratives. The method, however, depends heavily on the interviewer’s skill and may
introduce interpretive variation.

Unstructured Data Collection Methods

Unstructured methods provide maximum openness, enabling participants to express ideas in
an unrestricted manner. There is no fixed set of questions, and the dialogue evolves naturally.
Common formats include open-ended interviews, unstructured group discussions,
ethnographic observations, and narrative-based approaches.
Businesses adopt unstructured methods when exploring new concepts, understanding
consumer emotions, or identifying issues that are not yet clearly defined. For example, a
company developing an innovative product may use unstructured conversations with potential
users to uncover unmet needs and expectations. These methods produce rich qualitative
insights, though analyzing such data requires substantial time and interpretive expertise.

Structured methods focus on uniformity, semi-structured methods maintain guided flexibility,

and unstructured methods support open exploration. Each approach contributes uniquely to
business research depending on the depth and type of information required.
 SET II

Q.4)

Ans- Simple linear regression is a statistical approach used to analyze how one predictor
variable influences a single outcome variable. In R, this method is implemented through built-
in functions that estimate the slope, intercept, and overall relationship between the variables.
The process helps researchers understand trends and quantify the strength of association. In
business research, this model is applied to study patterns such as how advertising budget
affects revenue, how experience influences productivity, or how customer ratings relate to
sales performance.

Implementing Simple Linear Regression in R

The lm() function is the primary tool for fitting a simple linear regression model. It accepts a
formula that expresses the dependent variable as a function of the independent variable. After
fitting the model, users can extract coefficients, significance values, and performance
indicators through summary outputs.

Example implementation:

# Sample dataset
hours <- c(5, 7, 8, 10, 12, 15, 18)
scores <- c(55, 60, 65, 70, 74, 80, 85)

# Creating a data frame

df <- [Link](hours, scores)

# Fitting the regression model

model <- lm(scores ~ hours, data = df)

# Viewing model summary

summary(model)

The summary output provides essential results such as the estimated slope showing how
scores change with study hours, the intercept value, the significance of the model, and the R-
squared value that explains how well the independent variable accounts for variation in the
dependent variable. These indicators help researchers interpret whether the model holds
predictive relevance in practical scenarios.

Visualizing Simple Linear Regression in R

Visualization plays a critical role in confirming the pattern between variables and
communicating the findings clearly. Base R functions provide a straightforward approach for
generating scatterplots and overlaying regression lines.

# Scatterplot with regression line

plot(df$hours, df$scores,
main = "Study Hours vs Test Scores",
xlab = "Study Hours",
ylab = "Scores",
pch = 19)

abline(model, col = "blue", lwd = 2)

The scatterplot presents the observed data points, and the regression line illustrates the
predicted trend. This visual combination helps researchers assess linearity, identify potential
outliers, and evaluate how closely the model fits the data.

Using ggplot2 for Enhanced Visualization

For refined presentation and publication-quality output, the ggplot2 package offers advanced
tools for plotting regression results.

library(ggplot2)

ggplot(df, aes(x = hours, y = scores)) +

geom_point() +
geom_smooth(method = "lm", se = FALSE) +
labs(title = "Regression Analysis",
x = "Study Hours",
y = "Scores")

This visualization enhances clarity and supports clean documentation for academic or
business research reports, making regression results more interpretable for stakeholders.

Q.5)

Ans- A research report is a structured document that presents the entire journey of a study,
beginning with the research problem and ending with conclusions and implications. It serves
as a medium through which findings are communicated to readers in a clear, logical, and
evidence-based manner. In business research, the report supports informed decision-making
by translating data and analysis into actionable insights. A well-organized report enhances
credibility and allows others to understand how the study was conducted.

Title Page
The title page presents basic identifying information. It includes the report title, researcher’s
name, institutional association, and submission date. A clear and focused title enables readers
to grasp the central theme of the investigation at a glance.

Abstract or Summary
The abstract summarizes the core elements of the study, including the problem, objectives,
methods, and main findings. Business-oriented reports often use a brief executive summary to
highlight essential insights for managers who require quick interpretation without reviewing
the entire document.

Introduction
The introduction outlines the context of the study, defines the research problem, states the
objectives, and explains the relevance of the topic. It introduces the rationale behind the
research and sets the stage for the sections that follow. This part creates a foundation by
linking the study to broader business issues or industry concerns.

Review of Literature
The literature review discusses previous studies, theories, and existing knowledge related to
the topic. It identifies gaps or inconsistencies that justify the need for new research. This
section demonstrates familiarity with scholarly work and establishes the conceptual
framework guiding the investigation.
Research Methodology
The methodology section explains how the study was carried out. It describes the research
design, sampling methods, data collection tools, variables, and analytical techniques. This
part ensures transparency and allows readers to assess the reliability, validity, and
replicability of the study.

Data Analysis and Interpretation

Data analysis presents organized results in the form of tables, charts, graphs, or statistical
outputs. Interpretation explains the meaning of these results and connects them with the
research objectives. In business research, this section often emphasizes practical insights and
patterns that have strategic or operational significance.

Findings
The findings section highlights the key outcomes derived from the analysis. It summarizes the
most important results and presents them clearly so that readers can understand what the
study discovered.

Conclusion and Recommendations

The conclusion synthesizes the overall results. Recommendations provide practical guidance
rooted in the findings, helping organizations apply the insights to real business situations.

References and Appendices

References include all sources cited in the report. Appendices contain supplementary items
such as questionnaires, raw data, detailed tables, or methodological notes.

Q.6)

Ans- The objective is to examine whether gender and highest education level are independent
in the surveyed population of 395 respondents. Both variables are categorical, so a chi-square
test of independence is appropriate. The analysis uses a 5% significance level (α = 0.05).

Observed Frequencies
The contingency table from the survey is:

Gender High School Bachelors Masters Ph.D. Total

Female 60 54 46 41 201
Male 40 44 53 57 194
Total 100 98 99 98 395

Hypotheses

 Null hypothesis (H₀): Gender and education level are independent.

 Alternative hypothesis (H₁): Gender and education level are dependent.

Expected Frequencies
Expected counts under independence are computed as (row total × column total) / grand total.
Rounded expected values are:

 Female, High School: E = 201×100/395 = 50.886

 Female, Bachelors: E = 201×98/395 = 49.868
  Female, Masters: E = 201×99/395 = 50.377
 Female, Ph.D.: E = 201×98/395 = 49.868
 Male, High School: E = 194×100/395 = 49.114
 Male, Bachelors: E = 194×98/395 = 48.132
 Male, Masters: E = 194×99/395 = 48.623
 Male, Ph.D.: E = 194×98/395 = 48.132

All expected cell counts exceed 5, so the chi-square approximation is valid.

Chi-Square Statistic and Components

The chi-square contribution for each cell is calculated as (O − E)² / E. The eight contributions
(rounded) are:

 Female–High School: 1.632

 Female–Bachelors: 0.342
 Female–Masters: 0.380
 Female–Ph.D.: 1.577
 Male–High School: 1.691
 Male–Bachelors: 0.355
 Male–Masters: 0.394
 Male–Ph.D.: 1.634

Summing these values yields the test statistic: χ² = 8.006 (approximately).

Degrees of Freedom and Critical Value

Degrees of freedom for a contingency table are (rows − 1) × (columns − 1) = (2 − 1) × (4 − 1)
= 3. The critical value from the chi-square distribution at α = 0.05 and df = 3 is χ²₀.₀₅,₃ = 7.815
(approximately).

P-value
Using the chi-square distribution, the p-value associated with χ² = 8.006 and df = 3 is
approximately 0.0459.

Decision Rule and Conclusion

The decision rule compares the test statistic to the critical value or uses the p-value against α.
Since χ² = 8.006 > 7.815 and p-value ≈ 0.0459 < 0.05, reject the null hypothesis at the 5%
significance level. The data provide sufficient evidence to conclude that gender and highest
education level are not independent in this sample.

The statistical result indicates an association between gender and education attainment
categories in the surveyed population. Examination of observed versus expected counts
suggests notable deviations: females exceed expected counts in High School and Bachelors
categories, while males exceed expectations in Masters and Ph.D. categories. For
practitioners, this pattern may signal gender-related differences in educational attainment that
merit further investigation, for example through stratified analyses or studies that incorporate
socioeconomic covariates to explore underlying causes.