I have experiment leading to repeated measures 4-way ANOVA:
Factors:
- group: two groups (between)
- intervention: three interventions (within)
- timepoint: two measurements, one before and one after intervention (within)
- an additional condition with 2 factor levels (within)
Since a Four-Way ANOVA is very complicated, I would like to perform a Three-Way ANOVA. However, each factor is of interest, so none should be dropped.
Possibilities:
- I have read about the possibility of merging two factors. Does anyone have any literature recommendations on this? (see: Can you convert three-way ANOVA to one-way ANOVA?)
- I was thinking that a change score could be calculated between the measurements before and after the intervention (or measurement-post - measurement-pre). Could I reduce the model with this "Change-Score" by the factor timepoint (pre and post-intervention measurements)? I am not sure if this is an acceptable practice.
