Suppose I have an experiment with 5 groups and 5 DVs and intend to conduct MANOVA to determine if there are mean differences between groups,
1. What is the minimum sample size needed for each group with Power = .80 and Alpha = .05?
2. Could you provide the numbers needed for small, medium, and large effect sizes for MANOVA, and references (journal articles) to these conventions?
3. If I do not have equal sample sizes for all groups, can I assume that as long as each of the groups meets the minimum number, we are maintaining the minimum power? Otherwise, what are the implications and how can I migitigate them?