A researcher plans a study to detect a clinically meaningful difference in mean hemoglobin of 1 g/dL between two groups. The standard deviation is estimated at 2 g/dL. Using alpha = 0.05 and power = 80%, the calculated sample size per group is 64. If the researcher wants to detect a smaller difference of 0.5 g/dL instead, what happens to the required sample size?

• A It decreases to 32 per group
• B It remains the same at 64 per group
• C It increases to 128 per group
• D It increases to 256 per group ✓

Explanation

Sample size is inversely proportional to the square of the detectable difference. Halving the difference from 1 to 0.5 g/dL requires quadrupling the sample size (64 x 4 = 256 per group). This relationship is critical in study design: smaller effect sizes demand substantially larger samples to maintain the same power and significance level.

Reference: Methods in Biostatistics, 8th ed.

High-yield for: NEET PGINI-CETNExTFMGEUSMLEPLABMRCP

Written and medically reviewed by the StethoPrep medical team.