BAPSY Methodical & Stats Analysis 4 — Questions and Answers
Question 1: Which of the following is a non-parametric alternative to the independent-samples t-test?
- Wilcoxon signed-rank test
- Mann-Whitney U test (Correct answer)
- Kruskal-Wallis test
- Friedman test
Correct answer: Mann-Whitney U test
The Mann-Whitney U test is the non-parametric equivalent of the independent-samples t-test, comparing two unrelated groups using ranks.
Question 2: Regression toward the mean is most likely to threaten internal validity in which design?
- Randomized controlled trials
- Pre-test/post-test designs without a control group (Correct answer)
- Counterbalanced within-subjects designs
- Latin square designs
Correct answer: Pre-test/post-test designs without a control group
When participants are selected based on extreme scores, scores naturally move toward the mean on retest, threatening internal validity in pre/post designs without controls.
Question 3: Which validity type is threatened when a measure does not adequately sample all facets of a construct?
- Predictive validity
- Convergent validity
- Content validity (Correct answer)
- Discriminant validity
Correct answer: Content validity
Content validity concerns whether a measure comprehensively covers the full domain of the construct being assessed.
Question 4: In meta-analysis, publication bias refers to:
- Errors introduced by coding studies inconsistently
- The tendency for significant results to be published more than null results (Correct answer)
- Using different effect size metrics across studies
- Weighting studies by sample size only
Correct answer: The tendency for significant results to be published more than null results
Publication bias occurs because journals disproportionately accept significant findings, inflating meta-analytic effect size estimates.
Question 5: A researcher uses stratified random sampling. The primary advantage over simple random sampling is:
- It requires a smaller overall sample
- It guarantees representation of key subgroups (Correct answer)
- It eliminates the need for random assignment
- It reduces demand characteristics
Correct answer: It guarantees representation of key subgroups
Stratified sampling divides the population into subgroups and samples from each, ensuring proportional representation of important strata.
Question 6: The standard error of the mean (SEM) decreases when:
- Population variance increases
- Sample size decreases
- Sample size increases (Correct answer)
- The measurement scale changes
Correct answer: Sample size increases
SEM = σ/√n, so increasing sample size reduces the standard error, making sample means more precise estimates of the population mean.
Question 7: When a researcher randomly assigns participants to conditions, the primary purpose is to:
- Increase external validity
- Equate groups on known and unknown confounds at baseline (Correct answer)
- Ensure blinding of participants
- Maximize statistical power
Correct answer: Equate groups on known and unknown confounds at baseline
Random assignment distributes both known and unknown participant characteristics evenly across conditions, controlling for confounds.
BAPSY Methodical & Stats Analysis 4
Which of the following is a non-parametric alternative to the independent-samples t-test?