Fig. 5
From: The impact of iterative removal of low-information cluster-period cells from a stepped wedge design
Precision loss (top panel) and power (bottom panel) corresponding to designs obtained through the iterative removal process, indicated by the removal percentage of cluster-period cells, for the exchangeable model (solid line) with a within-period ICC of \(\rho =0.14\) and a standardised effect size of 0.25, and the discrete-time decay model (dotted line) with a within-period ICC of \(\rho =0.15\) and CAC of \(r=0.95\), and a standardised effect size of 0.35