Research Fundamentals: Definitions, Types, and Proposal Structure

TAKEAWAYS:

• Research is a systematic, logical inquiry to solve problems or generate new knowledge, classified into basic (theoretical) and applied (practical) types.
• A research proposal is a blueprint that outlines the problem, objectives, methodology, and expected outcomes—required for ethical approval and funding.
• Objectives must be SMART (Specific, Measurable, Achievable, Relevant, Time-bound) and derived from the research question.
• Quantitative and qualitative research differ in data types (numbers vs. words) and analysis methods (statistics vs. thematic analysis).
• Real-world applications include eSewa’s fraud detection (quantitative data analysis) and Pathao’s driver satisfaction surveys (qualitative research).
• Plagiarism and ethical violations (e.g., falsifying data) can invalidate research and damage credibility.

---

1. What Is Research?

Research is a structured process of investigating a problem to discover solutions, improve understanding, or test hypotheses. It involves:

• Observation: Collecting data (e.g., traffic congestion in Kathmandu).
• Analysis: Interpreting data (e.g., using statistical tools to identify peak hours).
• Conclusion: Drawing actionable insights (e.g., proposing new bus routes).

Types of Research

Research is classified based on purpose, scope, and methodology. The primary categories are:

Real-World Example: eSewa’s Fraud Detection

• Problem: Online payment fraud in digital transactions.
• Research Type: Applied + Quantitative
  • Method: Analyzing transaction logs (big data) to detect anomalies using machine learning.
  • Outcome: Reduced fraud by 30% (as reported in eSewa’s 2023 annual report).

---

2. Why Is a Research Proposal Needed?

A research proposal is a formal document submitted to:

• Secure funding (e.g., grants from Tribhuvan University).
• Gain ethical approval (e.g., for human subject studies).
• Define the scope of the research to avoid ambiguity.

Components of a Research Proposal

A well-structured proposal includes:

How to Formulate Research Objectives

Objectives must be:

1. Specific: Target a clear issue.
   • ❌ "Study mobile banking."
   • ✅ "Assess the adoption rate of Khalti’s USSD service among farmers in Kavrepalanchok."
2. Measurable: Use quantifiable metrics.
   • ✅ "Achieve 70% response rate in the survey."
3. Achievable: Feasible with resources.
   • ❌ "Survey all 3M Khalti users." → Too broad.
   • ✅ "Survey 500 users in 3 districts."
4. Relevant: Align with the research question.
5. Time-bound: Set deadlines.
   • ✅ "Complete data analysis by June 2025."

Worked Example: Ncell’s Network Coverage Research

• Problem: Poor 4G signal in hilly regions of Nepal.
• Objective: "Map 4G coverage gaps in 10 districts using drive tests and customer complaints by December 2024."
  • Why SMART?
    • Specific: Focuses on 4G and hilly regions.
    • Measurable: Drive tests provide signal strength data.
    • Achievable: 10 districts are manageable.
    • Relevant: Directly addresses customer pain points.
    • Time-bound: Deadline set for December 2024.

---

3. Data Generation Methods

Research collects data through primary (firsthand) or secondary (existing) sources. Below are 7 key methods:

---

4. Quantitative vs. Qualitative Research

Data Analysis Techniques

• Quantitative:
  • Descriptive statistics (mean, median).
  • Inferential statistics (regression, hypothesis testing).
  • Example: "Analyzing NEPSE’s stock price trends using moving averages."
• Qualitative:
  • Thematic analysis (identifying patterns in text).
  • Content analysis (coding social media posts).
  • Example: "Coding Pathao driver complaints to find common themes."

---

5. Research Ethics and Plagiarism

Ethical Considerations

• Informed Consent: Participants must know the study’s purpose and risks.
  • Example: "Ncell must inform users before collecting call data for research."
• Confidentiality: Protect participant identities.
• Avoiding Harm: Ensure no physical or psychological harm.
• Transparency: Disclose funding sources and conflicts of interest.

Plagiarism: Types and Consequences

Plagiarism is intellectual theft and includes:

1. Direct Plagiarism: Copying text verbatim without citation.
2. Self-Plagiarism: Reusing your own work without acknowledgment.
3. Patchwriting: Paraphrasing poorly (e.g., changing a few words).
4. Ideas/Infringement: Using someone else’s theory without credit.

Consequences:

• Academic: Failing the course, expulsion.
• Professional: Loss of job, damaged reputation (e.g., a researcher at TU losing funding).
• Legal: Copyright infringement lawsuits.

How to Avoid Plagiarism:

• Use quotation marks and citations (APA, Harvard styles).
• Paraphrase correctly (e.g., change structure and synonyms).
• Use plagiarism checkers (Turnitin, Grammarly).

---

6. The Role of Literature Review

A literature review is a critical summary of existing research on your topic. It:

• Identifies gaps in current knowledge.
• Justifies your research’s novelty.
• Provides theoretical framework.

How to Organize a Literature Review

1. Start Broad: Overview of the field (e.g., "Digital payments in developing countries").
2. Narrow Down: Focus on your topic (e.g., "M-Pesa’s impact on poverty in Kenya").
3. Compare Studies: Highlight contradictions or consensus.
4. Link to Your Research: Explain how your study fills a gap.

Example for Nepali Context:

"While studies on M-Pesa (Kenya) and GCash (Philippines) show 30–50% adoption rates, Nepal’s Khalti has only 20% adoption in rural areas (NIBL, 2023). This review explores whether Khalti’s USSD service—unlike app-based models—can bridge the digital divide."

---

7. Research Report Writing

A research report presents findings clearly and professionally. Key components:

Exam Tip: Use headings, bullet points, and visuals (tables, graphs) to make reports scannable. Avoid jargon.

---

In the Real World

1. eSewa’s Fraud Detection System

   • Idea Used: Quantitative research + machine learning.
   • How: eSewa analyzes transaction patterns (secondary data) to flag anomalies (e.g., sudden large transfers). Their 2023 report credits this with reducing fraud by 30%.
   • Link to Unit: Shows how data collection (secondary) and quantitative analysis solve real-world problems.

2. Pathao’s Driver Satisfaction Surveys

   • Idea Used: Qualitative research (interviews/focus groups).
   • How: Pathao conducts monthly surveys with drivers to identify issues like low pay or unsafe routes. In 2022, this led to a 15% pay hike for drivers in Kathmandu.
   • Link to Unit: Demonstrates research objectives (e.g., "Improve driver retention by 20% in 6 months") and methodology (surveys + thematic analysis).

3. Ncell’s 5G Rollout Strategy

   • Idea Used: Mixed-methods research (quantitative + qualitative).
   • How: Ncell combined:
     • Quantitative: Drive tests to measure 5G coverage in 10 districts.
     • Qualitative: Interviews with potential users (e.g., students, businesses) to gauge demand.
   • Outcome: Targeted 5G expansion in Pokhara and Dharan first, based on data showing higher smartphone penetration there.
   • Link to Unit: Illustrates SMART objectives ("Achieve 70% 5G coverage in 3 districts by 2025") and methodology (triangulation of data sources).

---

Exam Tip

1. For Short Notes (e.g., "Research Objectives"):

   • Use the SMART framework and give a Nepali example (e.g., "NTC’s objective to reduce internet outages by 50% in 2025").
   • Score Boosters:
     • Define SMART in your answer.
     • Compare with non-SMART objectives (e.g., "Study internet speeds" → vague).

2. For Long Questions (e.g., "Components of a Research Proposal"):

   • Structure: Use a table (like above) for components + examples.
   • Visuals: Include a timeline for methodology or a flowchart of the proposal process.

• Real-World Tie: Relate to eSewa/Khalti/Ncell (e.g., "Like eSewa’s fraud detection proposal, include a section on data privacy compliance").

3. Avoid Common Mistakes:

   • ❌ Generic examples (e.g., "Amazon" without linking to Nepal).
   • ✅ Localize: Use Nepali companies (Daraz, Pathao, NTC) or government data (NLSS, NIBL reports).
   • ❌ Ignoring ethics in proposals/reports.
   • ✅ Mention informed consent or confidentiality where applicable.

4. Diagrams That Score Marks:

   • Flowchart: Research process (problem → literature → methodology → report).
   • Table: Comparing quantitative vs. qualitative methods.
   • Graph: Presenting findings (e.g., "Ncell’s 5G coverage by district").

---

Final Checklist Before Submission:

• Used at least 2 real-world Nepali examples (e.g., eSewa, Pathao).
• Included visuals (table, flowchart, or image) for key concepts.
• Defined terms (e.g., SMART objectives, plagiarism types).
• Linked answers to exam question patterns (short notes vs. long answers).