“presize”: precision based sample size calculation
Sample size calculations are an important part of clinical research. Too few participants and the study objectives cannot be fulfilled; too many participants can be a waste of resources. Basically, there are two methods to calculate the sample size:
- Sample size calculation based on hypothesis testing (traditionally used)
- Precision based sample size calculation
The 2nd method is more in line with the current movement towards confidence intervals and away from p-values (reference to Martin Bland (BMJ 339).
While most statistical software packages contain some sample size calculation algorithms, they are mostly for method 1 based on hypothesis testing.
R is a statistical computing program and language commonly used in research. Statisticians at the CTU Bern, on behalf of the SCTO Statistics and Methodology Platform, have developed the “presize” R-package to assist with precision based sample size calculations. It includes methods for descriptive statistics, absolute and relative differences, various measures of correlation, as well as diagnostic measures.
“presize” is now available on R’s main package repository, CRAN. There is also a user-friendly web app, supporting almost the entire functionality of “presize”, for those who do not know R.
CRAN:
https://cran.r-project.org/web/packages/presize/index.html
Bug reports, feedback and feature requests:
https://github.com/CTU-Bern/presize/issues
Web app: