7 One-way ANOVA (between subjects)
7.1 Learning goals
7.1.1 SPSS
- Perform and interpret one-way ANOVA and post-hoc comparisons
7.2 Formulating hypotheses for one-way ANOVA
When comparing means of more than two independent groups, we move beyond t-tests and use Analysis of Variance (ANOVA). We formulate hypotheses as follows:
- \(H_0: \mu_1 = \mu_2 ... = \mu_k\) (where \(k\) = number of groups)
- Population means are equal across all groups.
- \(H_1:\) At least one of the population means of the \(k\) groups is different from (one of) the others.
Note that \(H_1\) here is not directional, and we use post-hoc tests to follow up on the omnibus F-test performed as part of the ANOVA procedure.
7.3 Explainer video (ISA mockup drugtrial-2)
- One-way ANOVA
See video ‘ISA7’ here