Table 4 Inference for the treatment effect, θ, and the within-period intracluster correlation, ρ1, for a randomly-selected simulated dataset for a SW design with S=1,T=5,m=10, true treatment effect θ=0, and intracluster correlation parameters ρ1=0.1 and r=1. Estimate and 95% CI correspond to point estimates and 95% confidence intervals for the REML methods and medians of posterior draws and 95% credible intervals for the Bayesian method. Note that standard errors for ρ1 are not provided in lme4 to permit 95% confidence intervals for the REML methods