Estimation and Testing of Wilcoxon–Mann–Whitney Effects in Factorial Clustered Data Designs

Clustered data arise frequently in many practical applications whenever units are repeatedly observed under a certain condition. One typical example for clustered animal experiments, where several animals share the same cage and should not be assumed to completely independent. Standard methods analysis of such Linear Mixed Models Generalized Estimating Equations—however, checking their assumptions is easy, especially scenarios with small sample sizes, highly skewed, count, ordinal or binary data. In situations, Wilcoxon–Mann–Whitney type effects suitable alternatives mean-based other distributional approaches. Hence, no specific distribution, symmetric asymmetric, required. Within this work, we will present different estimation techniques factorial designs discuss quadratic- multiple contrast type-testing procedures hypotheses formulated terms effects. Additionally, framework allows occurrence missing data: testing based on all-available instead complete-cases. An extensive simulation study investigates precision estimators behavior test type-I error control. real world dataset exemplifies applicability newly proposed procedures.