Research MethodologyUnit 311 min read
Research Questions: Types, Formulation & Validation
Unit 3 of Research Methodology explores how to identify meaningful research questions, classify them by type (exploratory, descriptive, explanatory, evaluative), and formulate them using PICO/T, while avoiding pitfalls like bias or vagueness. Includes real-world examples from Nepali tech companies (eSewa, Daraz) and st
Key Concepts and Definitions
What is a Research Question?
A research question is a clear, focused, and answerable query that guides your study. It defines the scope, purpose, and direction of your research. Unlike general questions, research questions are:
- Specific: Narrow enough to investigate thoroughly.
- Feasible: Achievable with available resources.
- Relevant: Aligns with the research problem and objectives.
- Ethical: Does not harm participants or violate norms.
Types of Research Questions
Research questions can be categorized based on the nature of the study and the type of information sought. The four primary types are:
| Type | Purpose | Example (Nepal Context) | Methods Used |
|---|---|---|---|
| Exploratory | To explore a new area or phenomenon. | "What are the challenges faced by freelancers in Nepal using digital platforms like Fiverr?" | Literature review, interviews, case studies |
| Descriptive | To describe characteristics or behaviors. | "What is the current adoption rate of digital wallets (eSewa, Khalti) among youth in Pokhara?" | Surveys, observations, statistical analysis |
| Explanatory | To explain relationships or causes. | "How does the use of mobile banking (NMB, Global IME) affect financial literacy in rural Nepal?" | Experiments, regression analysis, causal modeling |
| Evaluative | To assess the effectiveness of a program or policy. | "How effective is the NTC’s ‘Digital Nepal’ initiative in improving internet accessibility in remote areas?" | Program evaluation, surveys, impact assessment |
Formulating Research Questions
Step-by-Step Formulation
Formulating a strong research question involves breaking down the problem into manageable parts. Use the PICO(T) framework (common in medical and social sciences but adaptable to IT/research contexts):
| Component | Meaning | Example (Nepal IT Context) |
|---|---|---|
| P (Population) | Who or what is being studied? | "Small-scale farmers using mobile apps for agricultural advice in Kavrepalanchok." |
| I (Issue/Intervention) | What is the focus? | "Adoption of the ‘Kisan Suvidha’ app for weather forecasts and market prices." |
| C (Comparison) | What is being compared? (Optional) | "Compared to traditional radio/TV weather updates." |
| O (Outcome) | What are the expected results? | "Increased crop yield and reduced post-harvest losses." |
| T (Time) | What is the timeframe? (Optional) | "Over a 6-month period in 2024." |
Worked Example: Problem: Many students in TU fail to submit assignments on time due to poor internet connectivity. Research Question: "To what extent does unreliable internet infrastructure in TU campuses (e.g., Pulchowk, IOE) impact the timely submission of online assignments by Bachelor of IT students, and how does this compare to students using mobile data (Ncell/Jio)?"
Common Pitfalls and How to Avoid Them
Too Broad or Vague:
- ❌ "What is the impact of technology on education?"
- ✅ "How does the use of collaborative tools (Google Classroom, Microsoft Teams) improve group project outcomes for IT students at TU?"
- Fix: Narrow the scope using PICO(T).
Leading or Biased:
- ❌ "Does WhatsApp really improve communication efficiency in Nepali businesses?" (Assumes a positive answer)
- ✅ "How does WhatsApp Business API adoption affect customer response times in Nepali SMEs compared to traditional SMS?"
- Fix: Avoid loaded words like "really," "always," or "never."
Unanswerable or Unethical:
- ❌ "Why do some Nepali banks charge higher interest rates?" (Requires internal data access)
- ✅ "What are the perceived reasons among customers for choosing banks with higher interest rates on loans (e.g., NMB vs. Standard Chartered)?"
- Fix: Stick to observable, ethical data.
Validating Research Questions
A well-formulated research question must pass the following checks:
1. Feasibility Check
- Criteria: Can it be answered with available resources (time, budget, data)?
- Example:
- ❌ "What is the genetic history of all ethnic groups in Nepal?" (Requires DNA samples and years of study)
- ✅ "What are the perceived barriers to genetic testing among Newar communities in Kathmandu, based on a survey of 200 individuals?"
2. Originality Check
- Criteria: Does it add new knowledge or a unique perspective?
- Tools: Review literature to ensure the question hasn’t been fully answered.
- Example:
- ❌ "Impact of social media on mental health" (Overstudied)
- ✅ "How does the algorithm of TikTok’s ‘For You Page’ influence anxiety levels among Nepali teenagers aged 13–18, compared to YouTube Shorts?"
3. Ethical Check
- Criteria: Does it involve vulnerable groups or sensitive topics?
- Example:
- ❌ "Why do some Nepali women avoid using digital payment apps like Khalti?" (May require invasive interviews)
- ✅ "What are the perceived security concerns among women in Bhaktapur that prevent them from adopting mobile banking?"
In the Real World
eSewa and Research Questions:
- Idea Used: Evaluative Research Question
- Example: "How effective is eSewa’s ‘eSewa Pay’ feature in reducing cash transactions for utility bill payments (e.g., NTC, NEPSE) in Kathmandu, compared to traditional counters?"
- Why It Matters: eSewa uses such questions to refine their digital payment strategies based on user feedback and transaction data.
Daraz and Exploratory Questions:
- Idea Used: Exploratory Research Question
- Example: "What are the key factors influencing customer satisfaction with Daraz’s ‘Same-Day Delivery’ service in Pokhara, and how do these differ from Prime customers in the US?"
- Why It Matters: Daraz leverages exploratory studies to identify gaps in their logistics and customer service before scaling features.
Ncell and Descriptive Questions:
- Idea Used: Descriptive Research Question
- Example: "What is the current penetration rate of 4G services among rural households in Dhankuta, and how does it correlate with income levels?"
- Why It Matters: Ncell uses descriptive data to target marketing campaigns and infrastructure investments in underserved areas.
Worked Example: Kathmandu Traffic Routes
Scenario: The Kathmandu Metropolitan City (KMC) wants to reduce traffic congestion by optimizing bus routes. Research Question: "Which of the three proposed bus rapid transit (BRT) routes (Thapathali–Kageshwori, Thapathali–Naxal, or Thapathali–Kirtipur) would most effectively reduce travel time for TU students commuting from Bhaktapur, based on GPS data analysis over a 3-month period?"
Visualization of Routes:
Why This Matters:
- Type: Evaluative (assessing effectiveness of routes).
- Data Needed: GPS logs from student volunteers, traffic camera footage, and KMC’s existing route data.
- Outcome: Helps KMC prioritize infrastructure spending and aligns with TU’s sustainability goals.
Comparison: Qualitative vs. Quantitative Research Questions
| Aspect | Qualitative Research Questions | Quantitative Research Questions |
|---|---|---|
| Focus | Explores "why" or "how" with depth. | Measures "how much," "how many," or "to what extent." |
| Data Type | Words, narratives, themes (e.g., interviews, case studies). | Numbers, statistics (e.g., surveys, experiments). |
| Example (Nepal) | "How do Pathao drivers in Kathmandu perceive the impact of traffic police fines on their daily earnings?" | "What percentage of Pathao users in Lalitpur prefer cash-on-delivery over digital payments, and why?" |
| Tools | Thematic analysis, grounded theory. | Regression, chi-square tests, descriptive statistics. |
| Strengths | Reveals hidden motivations, rich context. | Generalizable, objective, scalable. |
| Weaknesses | Subjective, time-consuming, hard to generalize. | Lacks depth, may oversimplify complex issues. |
Exam Tip
Structure Your Answer:
- Start by classifying the research question (exploratory/descriptive/explanatory/evaluative).
- Use PICO(T) to break it down and show your understanding of scope.
- End with a validation (feasibility, originality, ethics).
Common Exam Pitfalls:
- ❌ Writing a question that is too broad (e.g., "Impact of technology on society").
- ❌ Forgetting to link the question to Nepal’s context (examiners love local relevance).
- ❌ Ignoring ethical considerations (always mention if sensitive groups are involved).
Scoring Boosters:
- Use real-world examples (e.g., eSewa, Daraz, Ncell) to illustrate your points.
- Compare qualitative vs. quantitative questions where applicable.
- Draw diagrams (like the PICO table or route flowchart) to visually explain complex ideas.
Based on the TU BIT syllabus for Research Methodology (RSM354), unit 3.
Discussion
Loading…