Figure 3From: Using Bonferroni, BIC and AIC to assess evidence for alternative biological pathways: covariate selection for the multilevel Embryo-Uterus model Power curves for the scenario where the prognostic variable under consideration affects the treatment at a uterus level and is tested in either the (incorrect) E or (correct) U sub-model. The power for a Bonferroni test with no sub-model assumption is included for comparison.Back to article page