Figure 2From: 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 an embryo level and is tested in either the (correct) E or (incorrect) U sub-model. The power for a Bonferroni test with no sub-model assumption is included for comparison.Back to article page