Skip to main content

Table 4 Simulated powers for a simple exchangeable correlation structure with the number of clusters = 8, cluster size = 120, and ICC = 0.1

From: Power calculation for detecting interaction effect in cross-sectional stepped-wedge cluster randomized trials: an important tool for disparity research

 

\({\theta }_{1}={\text{log}}\left(1.35\right)\)

\({\theta }_{1}={\text{log}}\left(1.68\right)\)

\({\theta }_{3}={\text{log}}\left(1.5\right)\)

\({\theta }_{3}={\text{log}}\left(2\right)\)

\({\theta }_{3}={\text{log}}\left(1.5\right)\)

\({\theta }_{3}={\text{log}}\left(2\right)\)

Prevalence rate of binary covariate

\(30\mathrm{\%}\)

\(50\mathrm{\%}\)

\(30\mathrm{\%}\)

\(50\mathrm{\%}\)

\(30\mathrm{\%}\)

\(50\mathrm{\%}\)

\(30\mathrm{\%}\)

\(50\mathrm{\%}\)

 GEE

0.834

0.876

0.999

1.000

0.839

0.882

0.999

1.000

 GEE-KC

0.769

0.817

0.996

0.998

0.775

0.824

0.996

0.999

 GEE-MD

0.697

0.748

0.989

0.995

0.702

0.756

0.989

0.995