Skip to main content
Fig. 1 | BMC Medical Research Methodology

Fig. 1

From: Confounding-adjustment methods for the causal difference in medians

Fig. 1

Bias in estimates of the causal difference in medians obtained under each method, for each skewness scenario and confounding setting in the simulated datasets (1000 datasets per skewness scenario). Scenarios 1 to 4 consider increasing levels of skewness, with Scenario 1 corresponding to the weakest and Scenario 4 corresponding to the highest skewness

Back to article page