Review of Literature: Definition and Importance

Definition of Review of Literature

Review of Literature (also called Literature Review) is a systematic and comprehensive examination of published and unpublished sources, research papers, books, journals, dissertations, and other scholarly materials related to a specific research topic or problem. It involves:

• Identifying relevant sources on the chosen research topic
• Analyzing and synthesizing existing knowledge
• Evaluating the quality and relevance of previous research
• Summarizing findings from prior studies
• Identifying gaps and limitations in existing research

A literature review presents a critical evaluation of what is already known about the research problem, rather than merely listing sources.

---

Importance of Review of Literature in Research Work

1. Establishes Research Foundation

• Provides background knowledge and context for the research problem
• Helps researchers understand the theoretical framework and existing concepts
• Builds a solid foundation upon which new research can be constructed

2. Identifies Research Gaps

• Reveals what has NOT been studied or explored
• Highlights areas where further investigation is needed
• Helps justify why the proposed research is necessary and original

3. Avoids Duplication

• Prevents researchers from repeating work already done
• Saves time and resources by learning from previous studies
• Ensures the research contributes something new to the field

4. Provides Methodological Insights

• Reveals research methods and approaches used by previous researchers
• Helps in selecting appropriate research design and methodology
• Identifies successful and unsuccessful approaches to similar problems

5. Supports Hypothesis Development

• Helps formulate research questions and hypotheses based on existing knowledge
• Provides theoretical basis for predictions and assumptions
• Strengthens the logical reasoning behind the research

6. Ensures Quality and Credibility

• Demonstrates that the researcher is knowledgeable about the field
• Adds credibility and validity to the research work
• Shows alignment with established academic standards

7. Identifies Key Variables and Concepts

• Clarifies important variables relevant to the research
• Defines key terms and concepts used in the field
• Helps in operationalizing variables for measurement

8. Facilitates Comparative Analysis

• Allows comparison of findings with previous research
• Helps identify trends, patterns, and contradictions in existing literature
• Enables contextualization of new findings within the broader field

---

Conclusion

A thorough review of literature is indispensable in research work as it provides the intellectual context, justifies the research need, guides methodology selection, and ensures that the research contributes meaningfully to the existing body of knowledge in the discipline.

Model Answer: Research Questionnaire and Its Requisites

Definition of Research Questionnaire

A research questionnaire is a structured data collection instrument consisting of a series of questions designed to gather information from respondents about their opinions, attitudes, experiences, beliefs, or behaviors. It is a systematic tool used in quantitative and qualitative research to obtain standardized responses that can be analyzed to answer research questions or test hypotheses.

Key characteristics:

• Comprises a set of predetermined questions
• Administered to a sample of respondents
• Responses are recorded in a standardized format
• Enables collection of large amounts of data efficiently
• Facilitates statistical analysis and comparison

---

Requisites of a Good Questionnaire

A well-designed questionnaire should possess the following essential qualities:

1. Clarity and Simplicity

• Questions must be clearly worded and easy to understand
• Avoid technical jargon, ambiguous terms, or complex sentence structures
• Use simple, direct language appropriate to the respondent's level

2. Relevance

• All questions must be directly related to the research objectives
• Each question should contribute meaningfully to answering the research questions
• Eliminate unnecessary or redundant questions

3. Validity

• Questions must measure what they are intended to measure
• Content should accurately represent the construct being studied
• Questions should have logical connection to research variables

4. Reliability

• Questions should yield consistent results when administered repeatedly
• Wording should be stable and not subject to different interpretations
• Responses should be reproducible under similar conditions

5. Objectivity

• Questions must be neutral and unbiased
• Avoid leading questions that suggest preferred answers
• Eliminate personal opinions or value judgments from question phrasing

6. Appropriate Length

• Questionnaire should be concise yet comprehensive
• Not too long to cause respondent fatigue or non-completion
• Balanced to collect sufficient data without being burdensome

7. Logical Organization

• Questions should follow a logical sequence
• Group related questions together
• Progress from general to specific or simple to complex topics
• Include proper transitions between sections

8. Appropriate Response Options

• Provide clear, mutually exclusive response categories
• Use consistent response scales throughout
• Include "Don't Know" or "Not Applicable" options where necessary
• Ensure options are exhaustive and non-overlapping

9. Anonymity and Confidentiality

• Protect respondent privacy
• Avoid asking for identifying information unless necessary
• Assure respondents of confidential treatment of responses

10. Pretesting

• Questionnaire should be pilot-tested before full administration
• Identify ambiguous questions or problematic wording
• Refine based on feedback from test respondents

---

Conclusion: A good questionnaire is the foundation of reliable data collection. It must balance comprehensiveness with brevity, clarity with completeness, and objectivity with relevance to ensure valid and reliable research findings.

Model Answer: Cluster Sampling

Definition of Cluster Sampling

Cluster sampling is a probability sampling technique in which the population is divided into naturally occurring groups or clusters, and then a random sample of clusters is selected. All elements within the selected clusters are included in the sample, or a further random sample is drawn from within each selected cluster.

Key characteristic: Sampling is done at the cluster level rather than the individual element level.

---

Conditions Where Cluster Sampling is More Suitable than Simple Random Sampling

1. Large Geographically Dispersed Population

• When the population is spread over a wide geographic area, cluster sampling reduces travel costs and time
• Example: Surveying households across multiple districts. Selecting entire villages (clusters) is more economical than randomly selecting scattered individual households
• Simple random sampling would require visiting dispersed locations, making it impractical and expensive

2. Population Lacks Sampling Frame

• When a complete list of all population elements is unavailable or difficult to obtain
• Clusters may be easier to identify and list than individual elements
• Example: Listing all students in a country is difficult, but listing schools (clusters) is feasible

3. Homogeneous Clusters with Heterogeneous Between-Clusters

• When elements within clusters are similar to each other, but clusters differ significantly from one another
• This ensures good representation of population diversity with fewer clusters needed
• Simple random sampling might miss important cluster-level variations

4. Cost and Administrative Efficiency

• Cluster sampling reduces administrative burden and data collection costs
• Interviewers/researchers can work in concentrated areas rather than traveling extensively
• Particularly suitable for large-scale surveys with limited budgets

5. Natural Cluster Structure Exists

• When the population naturally divides into clusters (schools, hospitals, factories, villages)
• Utilizing existing organizational structures is more practical than creating artificial sampling units

---

Summary

Cluster sampling is preferred over simple random sampling when cost efficiency, geographic convenience, and practical feasibility are priorities, especially for large, dispersed populations where a complete sampling frame is unavailable.

Model Answer: Plagiarism in Research & Stratified Random Sampling

a) Plagiarism in Research (2.5 marks)

Definition: Plagiarism is the act of presenting someone else's work, ideas, words, or intellectual property as one's own without proper acknowledgment or attribution. It is a serious academic and professional misconduct.

Key aspects:

• Forms: Direct copying of text, paraphrasing without citation, using others' research findings without credit, submitting work written by someone else
• Consequences: Academic penalties (failing grade, expulsion), damage to professional reputation, legal action in some cases, retraction of published work
• Prevention:
  • Always cite sources using proper referencing formats (APA, Harvard, IEEE, etc.)
  • Use quotation marks for direct quotes
  • Paraphrase with attribution
  • Maintain proper documentation of sources during research
  • Use plagiarism detection tools (Turnitin, etc.)

Ethical importance: Plagiarism violates academic integrity and intellectual property rights. Researchers must acknowledge the contributions of others to maintain credibility and trust in the scientific community.

---

b) Stratified Random Sampling (2.5 marks)

Definition: Stratified random sampling is a probability sampling technique where the population is divided into distinct, non-overlapping subgroups (strata) based on specific characteristics, and random samples are drawn from each stratum.

Key characteristics:

• Process:

  1. Divide population into homogeneous strata (e.g., by age, income, gender, location)
  2. Determine sample size for each stratum
  3. Randomly select samples from within each stratum

• Advantages:

  • Ensures representation from all subgroups
  • Reduces sampling error compared to simple random sampling
  • More precise estimates for population parameters
  • Useful when population has distinct groups

• Example: To survey student satisfaction in a university, divide students into strata (1st year, 2nd year, 3rd year, 4th year) and randomly sample from each year group proportionally.

• Types: Proportionate (sample size proportional to stratum size) and disproportionate (equal or weighted samples regardless of stratum size)