Unpaired t-test — MRCPsych Paper A MCQ

Instant feedback + full explanation. One question, done properly.

EasyEpidemiology and research methodsUnpaired t-testMRCPsych Paper A

A randomised trial compares mean 12-week depression rating-scale scores between two separate treatment groups. Individual scores are continuous, approximately normally distributed, and have similar variances in the two groups. Which statistical test is most appropriate?

Educational content. Not a substitute for clinical judgement or local policy.

Reveal the answer and explanation

Correct answer: B — Unpaired t-test

Explanation lettering: E = shown as A · C = shown as B · A = shown as C · B = shown as E

C is correct. An unpaired t-test compares the means of a continuous, approximately normally distributed variable between two independent groups. The stem also specifies similar group variances, satisfying the conventional Student unpaired t-test assumptions. A paired t-test is used when observations are linked, such as repeated measurements on the same participants or matched pairs. McNemar's test compares paired binary outcomes. A chi-squared test assesses associations or differences involving categorical frequency data rather than continuous means. Spearman rank correlation measures the strength of a monotonic association between two ranked, ordinal or non-normally distributed variables; it does not compare the means of two treatment groups.

Reference: Kim TK. T test as a parametric statistic. Korean Journal of Anesthesiology. 2015;68(6):540–546. https://pubmed.ncbi.nlm.nih.gov/26634076/