In calculating sample size for a cohort study comparing two proportions, which factor directly REDUCES the required sample size?

• A Increasing the desired power from 80% to 90%
• B Decreasing the acceptable Type I error from 0.05 to 0.01
• C Increasing the expected effect size (difference between proportions) ✓
• D Increasing the estimated prevalence of the outcome in the unexposed group

Explanation

Sample size for comparing two proportions is inversely related to the square of the effect size (difference between proportions). A larger expected effect size means a smaller sample is needed to detect it with adequate power. In contrast: increasing power (80% → 90%) increases sample size; decreasing α (0.05 → 0.01) increases sample size; increasing baseline prevalence in the unexposed group affects calculation but generally increases sample size when it moves the baseline proportion away from 0.5 (which maximizes variance). Effect size is the most practically important determinant of sample size.

Reference: Park's Textbook of Preventive and Social Medicine, 27th ed.

High-yield for: NEET PGINI-CETNExTFMGEUSMLEPLABMRCP

Written and medically reviewed by the StethoPrep medical team.