Paired t-tests and Wilcoxon signed rank tests (Chap. 2) require, just like multivariate data, two outcome variables, like the effects of two parallel treatments. However, they can not assess the effect of additional predictors like patient characteristics on the outcomes, because they have no separate predictor variables for that purpose. Generalized Linear Models can simultaneously assess the difference between two outcomes, and the overall effect of additional predictors on the outcome data.
CITATION STYLE
Cleophas, T. J., & Zwinderman, A. H. (2016). Paired Continuous Data with Predictors (Generalized Linear Models, 50 Patients). In SPSS for Starters and 2nd Levelers (pp. 11–15). Springer International Publishing. https://doi.org/10.1007/978-3-319-20600-4_3
Mendeley helps you to discover research relevant for your work.