Research Methodology — MCQ

Module 1

1. Which is NOT a characteristic of good research? a) Systematic b) Controlled c) Empirical d) Expensive

2. "Basic research" is mainly aimed at: a) Solving a specific problem b) Discovering new knowledge c) Testing a product d) Marketing

3. Belief that reality is socially constructed reflects: a) Positivism b) Post-positivism c) Constructivism d) Empiricism

4. Which philosophy holds that observation is theory-laden? a) Positivism b) Post-positivism c) Constructivism d) Realism

Module 2

5. A sharply defined research problem benefits from: a) Iterative refinement b) Ignoring feasibility c) No literature review d) A single broad question

6. In a fishbone diagram, the "head" is: a) A cause category b) The effect/problem c) The data d) The solution

7. The "5-Why" method is used for: a) Measuring sample size b) Root-cause analysis c) Randomisation d) Scale validation

Module 3

8. Which design can most strongly support a causal claim? a) Cross-sectional survey b) Correlational study c) Experiment with random assignment d) Case study

9. Randomising within a known block (e.g., sex) is: a) Simple random sampling b) Systematic sampling c) Stratified sampling d) Cluster sampling

10. H₁ stating "exercise reduces stress" (only stress decrease counts) is: a) Null hypothesis b) Two-tailed c) One-tailed (directional) d) Non-directional

11. Which is NOT a threat to internal validity? a) Confounding b) Selection bias c) History effects d) Reliability

Module 4

12. A 5-point "strongly disagree to strongly agree" scale measures: a) Nominal b) Ordinal c) Interval d) Ratio

13. Cronbach's α = 0.85 indicates: a) Poor reliability b) Acceptable reliability c) Excellent reliability d) Bad validity

14. Which is a type of validity, not reliability? a) Test-retest b) Internal consistency c) Construct d) Parallel forms

Module 5

15. In stratified sampling, sampling is done: a) Within each stratum b) Of whole strata only c) By convenience d) By snowball

16. The larger the sample size, the smaller the: a) Sampling frame b) Population c) Sampling error d) Bias

17. p = 0.5 is used for sample-size planning because: a) It gives the smallest n b) It gives the largest n c) It is always known d) It matches the mean

Module 6

18. For data with outliers, the better measure of centre is: a) Mean b) Median c) Mode d) All are equal

19. Pearson correlation r = −0.80 means: a) Strong positive linear relationship b) Weak negative relationship c) Strong negative linear relationship d) No relationship

20. In y = 3 + 2x, the slope means: a) y is always positive b) For each +1 in x, y increases by 2 c) y increases by 3 when x=0 d) x increases by 2

Module 7

21. P(A and B) for independent events equals: a) P(A) + P(B) b) P(A) − P(B) c) P(A) · P(B) d) P(A) / P(B)

22. Bayes' theorem lets you update: a) A mean b) A probability given evidence c) A standard deviation d) A sample size

23. The Central Limit Theorem mainly concerns the: a) Sample mean b) Population median c) Standard deviation d) Variance

Module 8

24. A p-value of 0.06 at α = 0.05: a) Reject H₀ b) Fail to reject H₀ c) Accept H₀ d) Invert H₀

25. Which test compares two independent group means (small samples)? a) Z-test b) Paired t-test c) Two-sample t-test d) Chi-square

26. The probability of a Type I error is: a) β b) α c) 1−α d) Power

Module 9

27. "Prep → Incubate → Illume → Verify" describes: a) The research process b) A creativity model c) The writing process d) The peer-review process

28. IEEE citation style uses: a) Author-date b) Numbered brackets c) Narrative citations d) Footnotes

Module 10

29. The Belmont Report emphasizes: a) Speed, cost, impact b) Respect, Beneficence, Justice c) Money, Fame, Publication d) Data, Code, Hardware

30. Fabricating data is: a) A reliability issue b) A validity threat c) Research misconduct d) A type of bias

Module 11

31. A CS research paper should demonstrate the artefact is: a) Only implemented b) Correct, analysed, and empirically evaluated c) Cheap and popular d) Written in LaTeX

32. For a simulation-based CS study, the biggest validity threat is: a) Low alpha b) Model realism c) Number of citations d) Page limit

33. In a CS proposal evaluation plan, the baselines provide: a) A literature review b) A comparison to beat c) Funding contacts d) A timeline

Answers

Reveal answers after attempting the questions

1-d, 2-b, 3-c, 4-b, 5-a, 6-b, 7-b, 8-c, 9-c, 10-c, 11-d, 12-b, 13-c, 14-c, 15-a, 16-c, 17-b, 18-b, 19-c, 20-b, 21-c, 22-b, 23-a, 24-b, 25-c, 26-b, 27-b, 28-b, 29-b, 30-c, 31-b, 32-b, 33-b.

Extended practice: Questions 34–70

No.	Question	A	B	C	D
34	Applied research primarily aims to:	Expand theory only	Solve a practical problem	Avoid data	Produce philosophy only
35	Quantitative research mainly emphasizes:	Numerical measurement and analysis	Narratives only	No hypotheses	Artistic interpretation
36	A research objective should be:	Vague and broad	Specific and assessable	Identical to the title	A result stated in advance
37	Feasibility commonly evaluates time, cost, access, skills, and:	Ethics	Font choice	Citation count	Popularity only
38	A fishbone diagram organizes:	Possible causes of a problem	Sample means	Citations	Random assignments
39	A hypothesis that states no difference is the:	Directional hypothesis	Null hypothesis	Alternative hypothesis only	Causal model
40	Random assignment primarily strengthens:	Internal validity	Citation validity	Face validity only	Population size
41	External validity concerns:	Generalizability	Measurement consistency only	Arithmetic accuracy	Plagiarism
42	Reliability means a measure produces:	Consistent results	The intended construct necessarily	Significant p-values	Large samples
43	Triangulation uses multiple methods/sources to improve:	Credibility	Random error always to zero	Sample size formula	Copyright
44	Temperature in Celsius uses which scale?	Nominal	Ordinal	Interval	Ratio
45	Weight in kilograms uses which scale?	Nominal	Ordinal	Interval	Ratio
46	A Likert item usually asks for:	Degree of agreement	A ratio measurement	A random number	An open biography only
47	Pilot testing is done mainly to:	Detect instrument problems before the main study	Replace sampling	Prove the hypothesis	Increase plagiarism
48	In simple random sampling, each unit has:	A known selection chance	No selection chance	Selection by researcher judgment	Selection by referrals
49	Stratified sampling draws samples from:	Defined subgroups	One natural cluster only	Volunteers only	A referral chain
50	Snowball sampling is useful for:	Hard-to-reach populations	Perfect population lists	Laboratory randomization	Census only
51	Sampling error arises because:	A sample rather than the full population is observed	Data are always fabricated	A citation is missing	Every instrument is invalid
52	The median is especially resistant to:	Extreme values	Sample size	Ordering	Categories
53	Standard deviation measures:	Spread around the mean	Causal effect	Sample identity	Scale validity
54	Correlation alone does not establish:	Causation	Association	Direction	Magnitude
55	In regression, the dependent variable is the:	Outcome being predicted	Predictor only	Sampling frame	Control group always
56	For mutually exclusive A and B, P(A∪B) equals:	P(A)+P(B)	P(A)P(B)	`P(A	B)`
57	The expected value of a discrete random variable is its:	Probability-weighted average	Largest value	Median always	Sample size
58	A normal distribution is symmetric about its:	Mean	Variance	Tail only	Range
59	Rejecting a true null hypothesis is a:	Type I error	Type II error	Power	Confidence level
60	Failing to reject a false null hypothesis is a:	Type I error	Type II error	Alpha level	Point estimate
61	Statistical power equals:	1−β	1−α	α+β	p-value
62	A paired t-test compares:	Two related measurements	Two unrelated large populations only	Categorical frequencies	Several variances only
63	Chi-square commonly tests:	Association between categorical variables	Two means only	A regression slope only	Normality of every value
64	An F-test/ANOVA comparison is based on a ratio of:	Variances	Medians	Sample sizes only	Citations
65	A literature review should primarily:	Synthesize and evaluate prior work	List sources without connection	Copy abstracts	Prove novelty by assertion
66	Plagiarism is using another's work without proper:	Attribution	Randomization	Tabulation	Scaling
67	A patent primarily protects a qualifying:	Invention	Brand name	Literary expression	Confidential consent form
68	Copyright primarily protects original:	Expression/work	Scientific fact	Research idea alone	Statistical test
69	Informed consent requires participants to understand the study and join:	Voluntarily	Secretly	Permanently	Without withdrawal rights
70	Fabrication differs from falsification because fabrication:	Invents data that never existed	Improperly manipulates existing research	Reuses text only	Omits a citation style

Extended answer key

34-B, 35-A, 36-B, 37-A, 38-A, 39-B, 40-A, 41-A, 42-A, 43-A, 44-C, 45-D, 46-A, 47-A, 48-A, 49-A, 50-A, 51-A, 52-A, 53-A, 54-A, 55-A, 56-A, 57-A, 58-A, 59-A, 60-B, 61-A, 62-A, 63-A, 64-A, 65-A, 66-A, 67-A, 68-A, 69-A, 70-A.