MANOVA
When several dependent variables belong in one analysis, and when they do not.
Slides · Dataset
About 30 minutes to complete
Enrol to open the slides and download the data.
What this module covers
when several dependent variables belong in one analysis, and when they do not; what MANOVA protects you from; assumptions including Box's M and multivariate normality; the menu path; reading the multivariate tests table and choosing among Pillai's trace, Wilks' lambda, Hotelling's trace and Roy's largest root; following up correctly, and the Bonferroni-adjusted univariate tests; discriminant analysis mentioned as the alternative follow-up; APA reporting.
By the end of it
- When several dependent variables belong in one analysis, and when they do not
- What MANOVA protects you from
- Assumptions including Box's M and multivariate normality
- Reading the multivariate tests table and choosing among Pillai's trace, Wilks' lambda, Hotelling's trace and Roy's largest root
- Following up correctly, and the Bonferroni-adjusted univariate tests
- Discriminant analysis mentioned as the alternative follow-up
The dataset
Health education trial, three outcome scores — Sigma teaching dataset
A teaching dataset of 180 cases, generated from a fixed seed so every figure in the module reproduces exactly.
- CodebookIncluded
- Dataset (CSV)Included
- SPSS syntaxIncluded