Detailed Summary of the PPT: "Module 2 - Literature Review"

1. What is a Literature Review?

A literature review is a structured summary of previous research on a specific topic. It
synthesizes key findings, highlights gaps, and identifies areas for further study.

Key Aspects of a Literature Review:

✔ Summarizes existing research – What is already known?

✔ Finds gaps – What is missing?
✔ Identifies trends – How has research evolved?
✔ Critically analyzes – Not just a summary, but an evaluation of studies.

🔹 Example:
If researching "How cloud computing affects small businesses," a literature review will:
✅ Collect research on cost benefits, security risks, and adoption challenges.
✅ Compare different studies to find patterns and contradictions.
✅ Identify gaps, such as "lack of research on small businesses in rural areas."

2. Research Question vs. Hypothesis

Aspect Research Question Hypothesis
A clear question that research A testable prediction about the relationship
Definition
aims to answer. between variables.
Guides the study and identifies Suggests an expected outcome that can be
Purpose
the problem. tested.
"How does online learning affect "Students who take online classes will score
Example
student performance?" lower than those in traditional classrooms."

3. How to Write a Literature Review (Step-by-Step Guide)

Step 1: Define Your Research Question

 Must be specific and researchable.

 Example: "How does cloud computing impact the business performance of SMEs in
India?"

Step 2: Search for Relevant Literature

Where to Find Studies?

 Google Scholar – General academic research.

 IEEE Xplore – Technology and engineering papers.
 JSTOR – Humanities and social sciences.
 Scopus – Peer-reviewed journals.
 PubMed – Medical research.

Advanced Search Techniques

 Use Boolean operators:

o "cloud computing" AND "business performance" – Both terms must
appear.
o "cloud computing" OR "cloud storage" – Either term may appear.
o "cloud computing" NOT "Amazon AWS" – Excludes AWS-related results.

Step 3: Evaluate Your Sources

✔ Check if it's peer-reviewed – Prefer academic journals.

✔ Recent publications – Prefer studies from the last 5 years.
✔ Author credibility – Is the researcher experienced?
✔ Methodology used – Are the findings based on strong evidence?

🔹 Example of Evaluating Sources:

Source Peer-reviewed? Findings Reliable?

Smith (2022) Yes Cloud computing improves SME efficiency. ✅ Yes
Blog article No General opinion on cloud computing. ❌ No

Step 4: Organize the Literature

🔹 Ways to Organize Research:

✔ By Theme – e.g., Benefits of cloud computing, Security challenges, etc.
✔ Chronological Order – Shows how research has evolved over time.
✔ Methodological Approach – Categorizes studies based on how they were conducted.

✅ Example Organization Table:

Theme Study Key Findings

Cost Benefits Smith (2022) Cloud computing reduces IT costs by 30%
Security Challenges Johnson (2021) Security concerns prevent adoption
Adoption Trends Lee (2023) AI in cloud computing is increasing
Step 5: Summarize and Compare Studies

Rather than summarizing each study separately, compare them.

✅ Example Synthesis:

 Smith (2022) found cloud computing reduces costs for SMEs.

 Johnson (2021) found security risks slow adoption.
 Conclusion: Businesses benefit from cloud computing, but security concerns need to
be addressed.

Step 6: Identify Research Gaps

✔ Look for missing aspects in research.

✔ Identify under-researched areas or contradictions.
✔ Find unanswered questions.

✅ Example Research Gap:

"Most studies focus on large enterprises, but few explore cloud adoption in small
businesses in rural India."

Step 7: Write the Literature Review

✔ Introduction – Introduce the topic and research question.

✔ Body – Organize studies into themes and compare findings.
✔ Conclusion – Summarize key insights, gaps, and future research directions.

✅ Example Conclusion:
"Cloud computing helps SMEs lower IT costs, but security risks are a major challenge.
Future research should explore how rural businesses can securely adopt cloud
technology."

4. Research Gap: What It Is and How to Find One

A research gap is an area that has not been fully explored.

How to Identify a Research Gap?

✔ Check if studies leave questions unanswered.

✔ See if research only focuses on certain groups.
✔ Look at the "Future Research" section of papers.
✅ Example of a Research Gap:
"Most studies on AI in healthcare focus on Western countries, but there is little research on
AI in Indian rural hospitals."

5. Types of Research Gaps

Type Definition Example
No theory explains a topic No model explains how AI affects
Theoretical Gap
fully. creativity.
Methodological Studies use outdated or weak Most research uses surveys, but no real-
Gap methods. world testing.
Research focuses only on Most studies on e-commerce focus on
Geographical Gap
certain regions. the US, not India.

6. What is a Hypothesis?
A hypothesis is a statement that predicts an outcome and can be tested.

Characteristics of a Good Hypothesis:

✔ Testable – Can be proven true or false.

✔ Specific – Focuses on one relationship.
✔ Based on Facts – Not a random guess.

✅ Example Hypothesis:
"Drinking more water will improve concentration in students."

7. Types of Hypotheses
Type Definition Example
No effect or "No difference in test scores between online
Null Hypothesis (H₀)
difference. and offline students."
Alternative A relationship "Online students score lower than offline
Hypothesis (H₁) exists. students."
Directional Predicts a specific
"More study time leads to higher test scores."
Hypothesis effect.

8. Hypothesis Testing Process

1️⃣ State the Hypothesis

 H₀ (Null Hypothesis): "No difference in test scores."

 H₁ (Alternative Hypothesis): "Online students score lower."

2️⃣ Choose a Significance Level (α)

 Usually 0.05 (5% chance of error).

3️⃣ Select a Statistical Test

 T-Test – Compares two groups.

 Chi-Square – Checks relationships between variables.

4️⃣ Analyze Data and Compare with α

 If p-value < 0.05, reject H₀ (significant difference exists).

✅ Example Result:
If p-value = 0.03, we reject H₀ → "Online students score lower."

Detailed Explanation of Hypothesis Testing Methods

The image lists five common hypothesis testing methods used in statistics. Here’s a
detailed explanation of each:

1. T-Test (Student’s T-Test)

📌 Used for: Comparing means of two groups when the sample size is small (n < 30) and
population standard deviation is unknown.

🔹 Types of T-Tests:

 Independent (Unpaired) T-Test: Compares means of two separate groups.

o Example: Checking if the average test scores of male vs. female students
differ.
 Paired T-Test: Compares means of the same group before and after an event.
o Example: Measuring students' math scores before and after tutoring.

2. Z-Test

📌 Used for: Comparing means when the sample size is large (n > 30) and the population
standard deviation is known.
🔹 Example:

 Comparing the average income of two cities when population variance is known.
 Checking if the average height of people in one country differs from another when
we have full population data.

✅ Difference Between T-Test & Z-Test:

 T-Test is used for small samples with unknown variance.

 Z-Test is used for large samples with known variance.

3. Chi-Square Test (χ² Test)

📌 Used for: Checking if two categorical variables are related.

🔹 Types of Chi-Square Tests:

 Chi-Square Test for Independence: Checks if two categories are associated.

o Example: Do gender and shopping preference have a connection?
 Chi-Square Goodness of Fit Test: Checks if observed data matches expected data.
o Example: Is the actual number of red, blue, and green M&Ms in a pack
close to the expected distribution?

✅ Used when data is in frequency counts (e.g., Yes/No, Male/Female, Pass/Fail).

4. ANOVA (Analysis of Variance)

📌 Used for: Comparing more than two groups to see if there is a significant difference
between them.

🔹 Example:

 Checking if the average salary differs among IT, Marketing, and HR departments.
 Comparing exam scores among students from three different schools.

🔹 Types of ANOVA:

 One-Way ANOVA: Compares one independent variable (e.g., salary across 3

departments).
 Two-Way ANOVA: Compares two independent variables (e.g., salary across 3
departments and based on gender).

✅ If more than two groups need to be compared, ANOVA is better than multiple T-
Tests!
5. MANOVA (Multivariate Analysis of Variance)

📌 Used for: Comparing more than one dependent variable across groups.

🔹 Example:

 Testing if different diets affect both weight loss and cholesterol levels.
 Checking if teaching methods impact both exam scores and class attendance.

✅ Difference Between ANOVA & MANOVA:

 ANOVA → Tests one dependent variable.

 MANOVA → Tests multiple dependent variables.

Detailed Explanation of Types of Hypotheses

The image lists different types of hypotheses. Let’s break them down with explanations and
examples for better understanding.

1. Simple Hypothesis

📌 Definition: A hypothesis that predicts a relationship between one independent

variable and one dependent variable.

🔹 Example:

 "Increasing the price of a product will reduce its sales."

(Only two variables: price and sales).

✅ Easy to test and analyze.

2. Complex Hypothesis

📌 Definition: A hypothesis that predicts a relationship between multiple independent

and dependent variables.

🔹 Example:

 "A healthy diet and regular exercise improve both heart health and mental well-
being."
(Two independent variables: diet, exercise; Two dependent variables: heart health,
mental well-being).
✅ Used in complex research studies.

3. Directional Hypothesis

📌 Definition: A hypothesis that specifies the direction of the relationship (positive or

negative).

🔹 Example:

 "Students who study more hours will score higher on exams."

(More study → Higher scores).

✅ Used when past research suggests a particular outcome.

4. Non-Directional Hypothesis

📌 Definition: A hypothesis that predicts a relationship exists but does not specify the
direction.

🔹 Example:

 "There is a difference in exam scores between students who study 5 hours and those
who study 10 hours."
(Does not specify if more studying improves or worsens scores).

✅ Used when the outcome is uncertain.

5. Null Hypothesis (H₀)

📌 Definition: A hypothesis that states there is no effect, relationship, or difference

between variables.

🔹 Example:

 "Drinking coffee has no effect on productivity."

✅ Used as a baseline for statistical testing. If the null hypothesis is rejected, an alternative
hypothesis is considered.

6. Alternative Hypothesis (H₁ or Ha)

📌 Definition: A hypothesis that states there is a significant effect, relationship, or
difference.

🔹 Example:

 "Drinking coffee increases productivity."

(If data supports this claim, we reject the null hypothesis).

✅ What researchers aim to prove.

7. Statistical Hypothesis

📌 Definition: A hypothesis that is tested using statistical methods to determine validity.

🔹 Example:

 "The average weight loss in a diet group is significantly different from the control
group."

✅ Analyzed using statistical tests like T-tests, ANOVA, or Chi-square tests.

8. Research Hypothesis

📌 Definition: A hypothesis formulated based on a research question and tested using

experiments.

🔹 Example:

 "Using digital learning tools improves students’ understanding of mathematics."

✅ Developed from real-world problems.

9. Associative Hypothesis

📌 Definition: A hypothesis that states one variable is associated with another, but it does
not prove causation.

🔹 Example:

 "People who exercise regularly tend to have lower stress levels."

(Exercise may be related to stress, but we don’t know if one causes the other).
✅ Correlation does not mean causation.

10. Causal Hypothesis

📌 Definition: A hypothesis that states one variable directly causes a change in another
variable.

🔹 Example:

 "Smoking causes lung cancer."

(Strong evidence that smoking is the cause).

✅ Requires experiments to prove causation (e.g., clinical trials).