Sample size calculation using G*Power software

Abstract

Sample size calculation is a critical step in planning a research project to obtain valid and generalizable results in biomedical sciences. It enables researchers to reach a balance between time, resources, and ethical considerations when planning and conducting research. Statistical software can assist in calculating sample size, but calculation method depends on the data characteristics and the research question.
This article presents a practical tool for calculating the required sample size for studies that aim to compare measures of central tendency for several research questions using the G*Power software. Sample size calculation for Student’s unpaired t-test, oneway analysis of variance (ANOVA), as well as their non-parametric analogs – Wilcoxon and Mann–Whitney tests are presented. The article walks readers step by step through sample size calculations for the abovementioned situations using practical examples. Therefore, this paper has a potential to become a valuable resource for junior researchers who are at the planning phase of their research. It serves as a practical guide for calculating the required sample size for parametric and nonparametric methods of analysis to test statistical hypotheses about the equality of measures of central tendency in unpaired groups.