Factorial ANOVAs Flashcards
What is a factorial ANOVA?
- Multiple independent variables
- Different PPs in all conditions (independent)
- One dependent variable
What do we find out from factorial ANOVAs?
How variables interact rathe than the main effect of the IV
What are the assumptions for a factorial ANOVA?
- independence
- normality
- homogeneity of variance
What do you use if the assumptions are not met?
- Non-parametric alternatives
- However, ANOVA is very robust so we can usually use them
What should you calculate when doing a factorial ANOVA?
- F ratio for the main effects
- Interaction effect
What is the use of a planned contrast?
You’re further dividing the variance explained by the model
What are custom contrasts?
- Used when you want to compare things in a way that built-in contrasts won’t allow
What are the rules of custom contrasts?
- Must be independent
- Only 2 chunks can be compared at once
- K-1 (should always end up with one less contrasts than the number of groups)
What are the two types of effect sizes?
- Eta Squared
- Partial Eta Squared
When do we use Eta Squared?
- One-Way ANOVAs
- Same as R^2
When do we use Partial Eta Squared?
- Factorial ANOVAs
- Tells proportion of variance that is uniquely explained by the IV