Biostatistics & Hypothesis Testing Flashcards

Which non-parametric test is the appropriate alternative to the independent samples t-test when data are not normally distributed?

Answer: Mann-Whitney U test

The Mann-Whitney U test compares the distribution of two independent groups using rank sums without assuming normality.

In survival analysis, censoring occurs when:

Answer: A subject's outcome is unknown after a certain time point

Censoring means the exact time-to-event is unknown because the subject left the study or follow-up ended without experiencing the event.

A researcher reports an odds ratio (OR) of 2.5 with a 95% CI of (1.1, 5.7). This result is:

Answer: Statistically significant and clinically meaningful

An OR of 2.5 with a 95% CI that does not include 1 (i.e., 1.1–5.7) indicates a statistically significant association at alpha=0.05.

Regression to the mean is most likely to bias results when a study selects participants based on:

Answer: Extreme baseline measurements

Participants selected for extreme baseline values tend to score closer to the population mean on re-measurement independent of any intervention.

In a 2×2 contingency table, the chi-square test assesses whether:

Answer: Two variables are independent

The chi-square test of independence evaluates whether two categorical variables are associated or independent of each other.

The coefficient of variation (CV) is useful because it:

Answer: Allows comparison of variability across variables with different units

CV = (SD / Mean) × 100% is a dimensionless ratio that allows meaningful comparison of variability between datasets measured on different scales.

A Kaplan-Meier curve that drops steeply early and then flattens indicates:

Answer: Most events occur early in follow-up

A steep early drop followed by a plateau means the majority of events (deaths, failures) happened soon after study entry.