Topic 8/9: Between subject Factorial Anova_Repeated Measure ANOVA Flashcards
Factorial design
More than 1 IV each with 2 or more levels
Between- subjects factorial ANOVA assumptions: (3)
- All groups are independent of eachother
- Populations are normally distributed
- Populations have equal variances
The Factorial ANOVA for a two factor design answers each of the following questions:
It does it by ….
- Is there a main effect of the first IV
- Is there a main effect of the second IV
- Is there an interaction effect between the two IVs?
By giving us an F-ratio for each of the questions that are asked
When you have alot of diff conitions
we get a smaller and smaller critical F to account for that
Null hypothesis/ alternative hypothesis for main effects and interactions
Assumption of repeated Measures ANOVA (3)
- Normally distributed data
- Homogenity of variance (Variance between Ps)
- Sphericity (variance between levels)
- Refers to the variances of the differences between all combinations of levels of the within-subjects factor
- Assumes that these differences are roughly equal
Repeated Measures ANOVA is robust to …. but not….
- violations of normality and homogenity
- sphericity
Sphericity example
- Marginal mean of treatment time are all roughly the same
Mauchly’s test of sphericity
- If p> 0.05 assume sphericity assumption met
- If p<0.05 sphericity assumption is violated
Hypothesis for repeated Measure ANOVA
Same as one way independent anova
Rejecting null for repeated measure ANOVA what is it phrased like?
At least one mean is different from another