Power and sample size calculations for a variety of study designs and outcomes. Methods include t tests, ANOVA (including tests for interactions, simple effects and contrasts), proportions, categorical data (chi-square tests and proportional odds), linear, logistic and Poisson regression, alternative and coprimary endpoints, power for confidence intervals, correlation coefficient tests, cluster randomized trials, individually randomized group treatment trials, multisite trials, treatment-by-covariate interaction effects and nonparametric tests of location. Utilities are provided for computing various effect sizes.

This package is a companion to the book, Crespi (2025) Power and Sample Size in R, Chapman and Hall/CRC. That book uses this package as well as many other R packages.

The latest stable version of powertools can be installed from CRAN using install.packages():

install.packages("powertools")

The current development version can be installed using devtools::install_github():

devtools::install_github("powerandsamplesize/powertools")

A study hopes to show whether or not a new experimental therapy is promising. A treatment is considered promising if at least 20% of participants respond well, and the researchers believe that the true response proportion is 30%. This is a basic example that shows how to calculate the sample size needed for this study to achieve 80% power, following Example 6.1 in the textbook.

library(powertools)
prop.1samp(N = NULL, p0 = 0.2, pA = 0.3, power = 0.8, sides = 1)
#> [1] 129.8337

prop.1samp returns N, the required sample size for a one proportion test.