Skip to main content


Fig. 4 | BMC Medical Research Methodology

Fig. 4

From: The impact of varying cluster size in cross-sectional stepped-wedge cluster randomised trials

Fig. 4

Impact of: a) coefficient of variation; b) cluster mean correlation; and c) number of clusters on the relative efficiency of a SW-CRT and parallel CRT, and impact of d) number of sequences on the relative efficiency of a SW-CRT. CV: Coefficient of variation. Efficiency is calculated as the ratio of the precision in a CRT with unequal cluster sizes compared to the precision in a CRT with equal cluster sizes. The green plots (with diamonds) are for a parallel CRT. The solid line indicates the median relative efficiency. The dashed lines indicate the lower and upper quartiles for the relative efficiency (short dashed line indicates SW-CRT). The default values for the design characteristics are (unless specified): number of clusters – 12; CV – 0.75; CMC – 0.2; number of sequences (SW-CRT only): 4

Back to article page