Which four factors should be considered when estimating sample size for a study?

• A. Frequency of disease, inter-individual variability, outcome difference between groups, Type I error
• B. Frequency of disease, inter-individual variability, outcome difference between groups, Type I error, Type II error
• C. Frequency of disease, inter-individual variability, random sampling, data normality
• D. Random sampling, data are interval, normal distribution, equal variances

Answer: (A)

The main idea behind estimating sample size is to match the study’s precision needs to what you expect to observe. Four core factors drive this planning: how common the outcome is, how much individual measurements vary, the size of the difference you want to be able to detect between groups, and how strict you want to be about avoiding a false-positive result. If the outcome is common, you have more information per participant and may need fewer subjects; if the outcome is rare, you’ll typically need a larger sample to observe enough events. Greater inter-individual variability reduces precision and pushes the required sample size up because each participant contributes more noise to the estimate. The difference between groups you aim to detect—your effect size—directly affects n: smaller differences require substantially more participants to demonstrate a statistically reliable separation. Finally, the Type I error level (alpha) sets how rarely you’re willing to claim a real effect when there isn’t one; lowering alpha makes it harder for results to reach significance and generally increases the needed sample size. The other options introduce factors that are more about study design assumptions or about other aspects of analysis rather than the principal determinants used in typical sample size formulas.