Essentials of sample size calculation

Abstract

Some essential information is required before the sample size of a study can be calculated. Choosing a study design is the first step. Next, we need to know the type of parameters of interest, inter-individual variability, precision, chance of errors, confidence level, anticipated drop-outs and number of covariates. While making an estimate of error, it is important to know whether the outcome can be one-sided or two-sided. Effect size should be chosen such that it is realistic, biologically plausible and of clinical significance. All this information should be gathered before calculating the sample size.

Keywords

• sample size
• study design
• error
• confidence level
• effect size