Skip to main content


Fig. 3 | BMC Medical Research Methodology

Fig. 3

From: Propensity score to detect baseline imbalance in cluster randomized trials: the role of the c-statistic

Fig. 3

Steps for bias detection and guidance for covariates adjustment. Our diagnosis tool corresponds to the top part of the graph (part 1), whereas the bottom part (part 2) is a qualitative approach to help to perform a covariate adjustment. Part 2 has to be thought in accordance to clinical knowledge about potential confounders. *Adjustment on predictors can increase precision in linear model and generally increases power in case of chance imbalance

Back to article page