topic Power calculation in Statistical Procedures

Power calculation

BayzidurRahman — Mon, 05 May 2025 07:49:29 GMT

Hi,

Can we calculate minimum detectable effect size for a cluster randomise trial(cRCT)? I have the following situation.

A two arm cRCT with 20 clusters in each arm, average cluster size is 30 (SD=20). Power=80%, Alpha=0.05. Proportion of outcome in the control group is 65%. What difference in the outcome between the 2 arms I can estimate from this study?

Bayzid

Re: Power calculation

Season — Mon, 05 May 2025 08:48:00 GMT

Take a look at this article: Sample size calculations for cluster randomised controlled trials with a fixed number of clusters | BMC Medical Research Methodology | Full Text.

An R package is also available: CRAN: Package CRTSize.

In addition, a book on this topic is available: Sample Size Calculations for Clustered and Longitudinal Outcomes in Clinical Research.

Re: Power calculation

BayzidurRahman — Mon, 05 May 2025 10:58:38 GMT

Hi,

Thanks for the references. Do you know of any SAS program or macro for this?

Re: Power calculation

Season — Mon, 05 May 2025 12:42:49 GMT

Unfortunately, I do not know. Maybe you can refer to the formula itself and try to compile codes on your own.

Re: Power calculation

BayzidurRahman — Thu, 08 May 2025 21:19:19 GMT

Hi,

Thanks for the references. I got the solution in Stata.