BDEsize: An R package for efficient determination of sample size in a balanced design of experiments

Abstract

The determination of an appropriate sample size in a balanced design of experiments is crucial for the statistical inference. The R package BDEsize computes the optimal sample size to detect a certain standardized effect size acquiring certain power at a significance level. Diagnostic graphs produced by the R package BDEsize provide the information of the mutual relationships between the sample size, the power, and the detectable standardized effect size. By investigating those graphs, we can identify some effects sensitive to the sample size, and adjust the sample size for achieving the purpose of experiment.

Keywords

• Sample size
• Balanced design of experiment
• Standardized effect size
• Power