Analysis of Variance Using SPSS and R

In this course a variety of analysis of variance (ANOVA) models is discussed, starting with a simple ANOVA but also including factorial designs and interaction effects, the inclusion of control variables, i.e., covariates (ANCOVA), and designs with within factors (repeated measures) and mixed between-within designs. The theory is presented in the mornings (lectures) and the afternoons are reserved for applications using both SPSS and R (computer lab).

The focus is not on the mathematics behind the statistical models but on applications in social (or other) sciences. Special attention will be paid for the problem of multiple testing and related issues like inflated Type I errors, alpha protection methods and subsequent reductions in power. On the last day of this course, an alternative for the classical ANOVA approach and subsequent post-hoc testing is presented. A non-technical introduction into Bayesian model selection for the evaluation of informative hypotheses is provided (including a computer lab in the afternoon where students practice with tailor-made software for this approach).

» Download the day-to-day programme (PDF)