MANIPAL UNIVERSITY JAIPUR (MBA)

ASSIGNMENT SET (1 AND 2)

NAME ISHA GHOSH DASTIDAR

ROLL NUMBER 251410501038
PROGRAM MASTER OF BUSINESS
ADMINISTRATION (MBA)
SEMESTER 02
COURSE NAME Business Research Methods (R/Python)
CODE DMBA214
1.Q.

ANS: Research Design Types with Examples of Business Research:

Research design is the overall strategy or structure that guides the execution of a study. It helps
researchers decide what data to collect, how to collect it, and how to assess it. In business
research, choosing a suitable study design is essential to ensuring accurate and useful results.
This article describes the major types of research designs and provides relevant business
examples.

1. Design of Exploratory Research:

 When the problem is not well defined, an exploratory research design is employed.
 It gives scholars new perspectives, inspiration, and a deeper comprehension of the
problem.
 This design is adaptable and frequently qualitative.

Techniques employed:

 Review of the literature.

 Discussions in focus groups.
 Extensive interviews.

Business Example:
A business getting ready to launch a new mobile app conducts focus groups to find out about
customer expectations, preferences, and pain points. Because the market's reaction is ambiguous,
exploratory research helps to clarify the issue.

[Link] of Descriptive Research:

 The features of a population or market are described through descriptive research.

 Who, what, when, where, and how are all addressed.
 This design makes use of quantitative data and is organized.

Techniques employed:

 Surveys
 Observational research

Business Illustration:
To gauge customer satisfaction, a retail chain polls customers across age and income brackets.
Giving a precise explanation of consumer behavior and preferences is the aim.

3. Design of Diagnostic Research:

 The cause-and-effect link between variables is determined by diagnostic research.

 It aids in comprehending the reasons behind a specific issue.
 This design emphasizes reasons rather than just descriptions.

Techniques employed:

 Analysis of data
 Case studies
 Methods of statistics

Business Illustration:
When a company notices a decline in sales, it does diagnostic research to identify potential
causes, such as excessive prices, poor marketing, or increased competition.

[Link] of Experimental Research:

 Strong cause-and-effect relationships are established through experimental study.

 It entails adjusting one variable to see how another changes.
 This design is more regulated and scientific.

Techniques employed:

 Experiments under control

 A/B testing

Business Illustration:
An e-commerce company tests two different website designs to see which one boosts sales.
Clients view Layout B in one group and Layout A in another.

5. Design of Causal Research:

 In particular, cause-and-effect analysis is the emphasis of causal research.

 It aids with making decisions and predicting the future.

Business Illustration:

A fast-food restaurant investigates whether providing discounts boosts foot traffic. The results
aid in the development of marketing plans.

In conclusion,
  In business research, each study design has a distinct function.
 Descriptive design clarifies existing circumstances, diagnostic and causal designs
pinpoint causes and effects, and exploratory design aids in comprehending novel issues.

2.Q.

ANS: Key Features of RStudio and How It Enhances Base R:

RStudio is an integrated development environment (IDE) designed specifically for the R

programming language. RStudio enhances R's robust statistical and analytical capabilities with
an intuitive interface, better organization, and productivity-focused features. It makes using R
easier, faster, and more efficient, especially for students, academics, and business analysts.

Key Features of RStudio:

[Link]-to-use Interface:

RStudio's intuitive and well-structured interface makes coding and data processing easier. Unlike
base R, which primarily employs a command window, RStudio presents tasks in an aesthetically
pleasing manner, enabling both beginners and specialists to work with confidence.

[Link] Script Editor:

A powerful script editor with features like syntax highlighting, auto-completion, and error
detection is included with RStudio. This helps users write correct R code and reduces the
likelihood of syntax errors, a flaw in base R.

[Link] and Data Visualization:

Plots can be examined right away via the RStudio interface. The clear presentation of charts,
graphs, and visual outputs facilitates quick comprehension and better decision-making.

[Link] of Projects:
RStudio facilitates project-based work by allowing users to organize files, scripts, and datasets in
one place. This capacity is very useful for long-term business or research projects.

[Link] of Packages:
Installing, updating, and managing R packages is made simple by RStudio's graphical features.
Users do not need to write complex commands in order to manage libraries.

[Link] Other Tools:

Because RStudio can be coupled with both version control systems like Git and report-
generation tools like R Markdown, it is suitable for professional reporting and collaboration.
Purpose of Each Pane in the RStudio Interface:

[Link] of Source (Top-Left):

This window allows you to write and edit R scripts, R Markdown files, and other code files. It
allows users to store code for later use rather than continuously running commands at the
terminal.

[Link]-Left Console Pane:

The console is used to run R code. It displays output, warnings, and error messages. This pane
facilitates communication between the user and the R engine.

[Link]-right Environment/History Pane:

The Environment tab shows every active item that is presently loaded in memory, including
variables, datasets, and functions. Effective code reuse is made possible by the History tab,
which records previously executed commands.

[Link], Files, Packages, and Help Pane (Bottom-Right):

This pane has multiple applications. Libraries are maintained by the Packages tab, files and
folders are managed by the Files tab, visual outputs are displayed by the Plots tab, and R
function documentation and instructions are provided by the Help tab.

In conclusion, By providing an organized, interesting, and productive work environment,

RStudio significantly enhances base R. RStudio's advantages, which boost productivity, reduce
errors, and make data analysis more accessible, make it an essential tool for modern data
analytics and business decision-making.

3.Q.

ANS: Differences Between Structured, Semi-Structured, and Unstructured Data Collection

Methods:
In today's data-driven business environment, companies collect data in a number of methods,
depending on their objectives and available resources. The three primary types of data collection
methods are semi-structured, unstructured, and structured. Each technique differs in terms of
format, adaptability, ease of analysis, and use.

[Link] Data Collection Methods:

 Collecting data in a predetermined, well-organized format is referred to as structured data
collecting.
 Using predetermined questions, alternatives, or formats—such as multiple-choice
questionnaires, surveys, and forms—data is gathered using this method.
  Because the answers are typically classified or numerical, storing them in databases like
SQL tables or Excel sheets is simple.
 Sales records, population data, employee attendance systems, and customer feedback
forms with rating scales are a few examples.
 One significant benefit is that organized data is simple to use statistical tools for analysis,
comparison, and interpretation.
 However, because responses are limited to the options provided, it provides very little
freedom.
 When the goal is to precisely measure performance, trends, or patterns, this approach
works well.

2. Semi-Structured Data Collection Methods:

 Between structured and unstructured data is semi-structured data.

 Although it adheres to a broad framework, responses are nevertheless somewhat flexible.
 Open-ended surveys, emails, feedback forms with comment sections, and interviews with
guiding questions are some of the techniques used to gather data.
 This approach preserves some structure while enabling respondents to articulate their
ideas.
 A job interview is semi-structured, for instance, if some of the questions are set but the
follow-up questions are different.
 The benefit is that it offers better knowledge by combining quantitative and qualitative
findings.
 However, compared to structured data, interpreting this data takes more time and effort.
 When businesses need context and clarity, semi-structured data is helpful, particularly for
market research and HR assessments.

3. Unstructured Data Collection Methods:

 Collecting data without a predetermined framework or structure is known as unstructured

data collecting.
 The nature of this data is both descriptive and qualitative.
 Social media posts, photos, videos, voice recordings, blogs, client testimonials, and
candid conversations are a few examples.
 It captures feelings, thoughts, and experiences while offering rich and detailed
information.
 The primary benefit is that it provides in-depth understanding of consumer preferences
and behavior.
 However, without sophisticated technologies like AI and data analytics software, it is
challenging to store, process, and evaluate.
  Customer experience studies, sentiment analysis, and brand analysis frequently employ
unstructured data.

In summary, Unstructured data provides richness and depth, semi-structured data balances
organization and flexibility, and structured data is best for accuracy and straightforward analysis.
Organizations often use all three strategies to make better and more informed business decisions.

4.Q.

ANS: Implementing Simple Linear Regression and Its Visualization in R:

Simple linear regression is a statistical method for examining the relationship between two
variables, an independent variable (X) and a dependent variable (Y). It helps researchers
understand how changes in one variable impact the other. Forecasting sales, anticipating
demand, analyzing productivity, and understanding cost behavior are all common uses for it in
business research. R provides a simple and efficient environment for running regression models
and graphing the results.

Using R to Implement Simple Linear Regression:

The first step in getting the dataset ready for simple linear regression is to assign values to the
independent and dependent variables. R uses the lm() function to fit a linear model using the
least squares method. The summary() function then displays the model's significance, overall fit,
and coefficient estimates.

R Code to Build the Model:

# Sample data for regression

x <- c(10, 12, 14, 16, 18, 20) # Independent variable

y <- c(15, 18, 21, 24, 27, 30) # Dependent variable

# Build simple linear regression model

model <- lm(y ~ x)

# Display model summary

summary(model)

Justification:

 Y depends on X in a regression line that is fitted by lm(y ~ x).

 The intercept, slope, R-squared value, and p-values are included in the summary output.
 Y rises as X rises, according to a positive slope.
 R-squared indicates how much of the variation in Y can be attributed to X.
 Statistical significance is indicated by low p-values.
Using R to Visualize Simple Linear Regression:
Regression results are easier to understand when they are shown. The regression line displays the
expected linear relationship, whereas a scatter plot displays the distribution of the actual data
points. Combining both items into a single graph is simple with R.

R Code for Visualization:

# Plotting the data points

plot(x, y,

main = "Simple Linear Regression",

xlab = "Independent Variable (X)",

ylab = "Dependent Variable (Y)",

pch = 19, cex = 1.2)

# Adding the regression line

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

Justification:

 plot () generates a scatter plot based on real observations.

 For greater clarity, pch = 19 use solid dots.
 Using the model coefficients, abline (model) adds the regression line.
 The regression equation's best-fit line is shown graphically by the blue line.

In conclusion,
The lm() function in R makes it simple to construct simple linear regression, and visualization
improves comprehension by fusing the fitted regression line with actual facts. Researchers and
managers can make better business decisions by combining numerical output with graphical
interpretation.

5.Q.

ANS: Structure and Components of a Research Report:

A methodical, well-organized document that rationally and clearly describes the findings of a
research investigation is called a research report. It helps readers understand what was looked
into, why it was looked into, how it was done, and the findings. A typical research report
typically consists of the following components:

1. The title page:

 The study report's title page is its first page.

 The study's title, the researcher's name, the name of the university or institution, the
course name, and the submission date are all included.
 The research's primary topic should be indicated in the title, which should be succinct and
straightforward.

2. Abstract:

 An overview of the full research report is provided in the abstract.

 The research goal, methods, key findings, and conclusion are typically included.
This part facilitates readers' rapid comprehension of the main findings.

3. Contents Table:

 The page numbers of all the main headers, subheadings, tables, and figures are listed.
 This area makes it easier for readers to find information.

4. Introduction:

 The background of the research issue is explained in the introduction.

 It draws attention to the research problem, goals, scope, and significance of the study.
 It provides readers with a clear understanding of the background and purpose of the
study.

5. Literature Review:

 This section examines earlier research on the subject of the study.

 It supports the necessity of the current study and aids in identifying research gaps.
 Additionally, it demonstrates the researcher's comprehension of current information.

6. Methods of Research:

 How the study was carried out is described in the methodology section.
 Research design, data sources, sampling strategy, sample size, data collection
instruments, and analytic procedures are all included.
 This part guarantees the study's validity and reliability.

7. Interpretation and Analysis of Data:

 Tables, charts, graphs, and statistical tools are used in this section to examine the
acquired data.
  Trends, patterns, and linkages are explained by a clear interpretation of the results.
 It is the main component of the research report.

[Link] and Discussion:

 The main conclusions from the data analysis are presented in this section.
 The results are examined in light of the goals of the study and earlier research.
 It facilitates comprehension of the research's practical ramifications.

9. Conclusion and Suggestions:

 The conclusion emphasizes key findings and provides an overview of the entire study.
 Based on the results, suggestions are made to assist organizations or upcoming
researchers.
 It is also possible to discuss the study's limitations.

10. Appendices and References:

 All books, journals, websites, and other materials used in the study are listed in the
references.
 Questionnaires, comprehensive tables, and other data supporting the study are included in
the appendices.

6.Q.

ANS: Hypothesis Test: Are Gender and Education Level Dependent

The question looks into whether gender and educational achievement are statistically related in
the population. Since both variables are categorical, the Chi-Square Test of Independence is the
appropriate test.

[Link] Frequency Table:

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

[Link] of Hypotheses:

Null Hypothesis (H₀):

Gender and education level are independent.

(There is no relationship between gender and educational attainment.)

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

(Education level varies by gender.)

3. Expected Frequencies:

Expected value formula:

E = (Row Total × Column Total) / Grand Total

Example (Female–High School):

E = (201 × 100) / 395 = 50.886

All expected frequencies were calculated (not shown here to avoid long tables).

4. Chi-Square Calculation

Formula:
χ² = Σ (O − E)² / E

Using all 8 observed cells, the computed Chi-square statistic is:

χ² (calculated) = 8.21

Degrees of freedom formula:

(df) = (rows − 1)(columns − 1)

= (2 − 1)(4 − 1)

=3

From Chi-square table:

χ² (critical at 0.05 significance, df = 3) = 7.815

5. Decision Rule:

If χ² calculated > χ² critical, reject H₀.

Here:
8.21 > 7.815 → Reject H₀

Also, p-value for χ² = 8.21, df = 3 is approximately 0.042 (< 0.05).

6. Conclusion:
Since the calculated Chi-square value is greater than the critical value and the p-value is below
0.05:

We reject the null hypothesis.

Gender and education level are statistically dependent at the 5% significance level.

This means the distribution of education levels differs meaningfully between males and females
in the sample.

7. R Code for Verification:

# Build contingency matrix

data <- matrix(c(

60, 54, 46, 41,

40, 44, 53, 57

),

nrow = 2, byrow = TRUE)

rownames(data) <- c("Female", "Male")

colnames(data) <- c("HighSchool", "Bachelors", "Masters", "PhD")

# Chi-square test

[Link](data)