Fig. 4From: Selecting a randomization method for a multi-center clinical trial with stochastic recruitment considerationsBoxplots of efficiency (relative to the idealized balanced design) in estimating the treatment effect assuming a linear model without covariates (Trial level, \({RE}_{1}\)), a linear model with region as a covariate (Region level, \({RE}_{2}\)), and a linear model with center as a covariate (center level, \({RE}_{3}\)) for 16 randomization designs under Scenario 1Back to article page