Fig. 3From: An imbalance in cluster sizes does not lead to notable loss of power in cross-sectional, stepped-wedge cluster randomised trials with a continuous outcomePower of stepped-wedge cluster randomised controlled trials (SW-CRTs) with varying average cluster size as variability in cluster size increases. The simulated power, relative to the analytical power, of SW-CRTs with increasing variability in cluster size, average cluster size ranging from 10 to 40, number of steps fixed at four and sample size calculated using the Woertman et al. design effect (DE) [7]. The type I error, power, intracluster correlation coefficient (ICC) and effect size were 0.05, 80%, 0.05 and 0.2, respectivelyBack to article page