Research Methodology — Important Questions

Answers below each question show the points expected in an exam.

Module 1

Short

1. Define research. State four characteristics of good research.
2. Distinguish basic and applied research with one example each.
3. What are the key tenets of positivism and constructivism?

Long

Discuss the evolution of scientific research from Bacon to the present day.

Answer points: Bacon's empirical method (induction) → Descartes' rationalism → Galileo's experiment → Newton's Principia (math + law) → 18th–19th c. specialisation and peer review → 20th c. statistics (Fisher) + Big Science → 21st c. data-intensive/open science/AI-assisted.

Module 2

Short

1. What is a research problem? Why must it be sharply defined?
2. State the five feasibility dimensions.

Problem

Refine "social media and mental health" into a research problem.

Answer: add population (e.g., 18–25 university students), intervention/exposure (e.g., >3 h daily use), outcome (anxiety/depression scores), and a time frame; state as a focused question, e.g., "Does >3 h/day social media use predict anxiety scores among 18–25-year-old students after 6 months?"

Module 3

Short

1. Differentiate experimental and non-experimental research design.
2. What is the role of randomisation and a control group?
3. Define internal and external validity.

Long

Explain validity, reliability, and triangulation. Why is each important?

Answer: validity = measuring/doing the right thing (construct/internal/external); reliability = consistent/stable (test-retest, α); both needed; triangulation uses multiple methods/sources to strengthen validity (reliability ≠ validity; validity requires at least acceptable reliability).

Module 4

Short

1. Give the four scale types and one permissible statistic for each.
2. What is Cronbach's α and when is it acceptable?
3. Distinguish construct and content validity.

Problem

Design a 4-item Likert scale for "library satisfaction" and state how you would test its reliability.

Answer points: 4 items ("The library is quiet enough," etc.) 1–5 agree scale; sum to a score; compute Cronbach's α from pilot data; α ≥ 0.7 acceptable.

Module 5

Short

1. Name two probability and two non-probability sampling methods.
2. What is sampling error and how is it reduced?
3. When is a sample size of 384 needed (95% confidence)?

Long

Explain stratified sampling and why it can be more precise than simple random sampling.

Answer: divide population into strata (homogeneous within, heterogeneous between), sample randomly within each stratum, combine. More precise because it guarantees representation of each stratum and reduces variance of estimates (especially if strata differ in the measured variable).

Module 6

Short

1. What is the difference between mean and median when data is skewed?
2. Define Pearson correlation; state its range and what r = 0 means.
3. In y = a + bx, what does b represent?

Problem

Given the data, compute the Pearson correlation. (Provide small numeric pairs.) Answer points: compute x̄, ȳ; compute r = Σ((x−x̄)(y−ȳ))/√(Σ(x−x̄)²·Σ(y−ȳ)²).

Module 7

Short

1. State the addition rule for non-mutually-exclusive events.
2. Give Bayes' theorem.
3. What does the Central Limit Theorem state?

Problem

A die is rolled twice. What is P(sum = 7)? Answer: 6 favourable pairs of 36 → 1/6.

Module 8

Short

1. Define Type I and Type II errors.
2. When do you use a Z-test vs a t-test?
3. What is the decision rule for p < α?

Long

A researcher reports t(24) = 2.30, p = 0.030 for a two-tailed test at α = 0.05. Interpret.

Answer: significant at 0.05 (0.030 < 0.05), reject H₀; moderate sample (df=24); two-tailed so effect in either direction; report effect size and CI too.

Module 9

Short

1. State Wallas' four stages of creativity.
2. What should a literature review end with?
3. Name four IPR types and what each protects.

Problem

Write an IMRaD outline for a paper on "predictors of student stress."

Answer points: I (gap, aim), M (design, participants, measures, analysis), R (results, no interpretation), D (meaning, limitations, implications).

Module 10

Short

1. State the three Belmont principles.
2. What are the FFP forms of research misconduct?
3. What must informed consent contain?

Long

Why is data minimisation important, and how does GDPR relate to research?

Answer: collect only necessary data to reduce privacy risk and comply with GDPR's data-minimisation principle; researchers must anonymise, get consent/specific conditions, and have a lawful basis (consent or public task); retain then delete.

Module 11

Short

1. What three elements should a strong CS paper address?
2. Distinguish experimental and simulation-based CS research.
3. What four elements belong in a proposal's evaluation plan?

Long

Outline a CS research proposal for improving battery life in mobile apps.

Answer: problem (battery drain), gap (existing profilers are coarse), approach (a static analyser flagging high-drain API use), evaluation plan (dataset of 50 apps, baselines [state of the art], metrics [battery % saved], statistics [paired t-test]); contribution + 6-month timeline + risks.

General (full-syllabus)

1. Trace the path from a research problem to an answer, naming the key stages.
2. Explain why randomisation matters more than blinding in a student-led survey.
3. Argue whether significance implies importance (include effect size).