SD6. Factorial ANOVA (repeated measures & mixed designs)

What does a two way ANOVA tell us?

main effects and interaction effects

What does an interaction tell us?

the dependency of one factor on another

What doesn't contribute to variance in repeated measures design?

individual differences because each ppt acts as their own control

What is the F ratio for two way repeated measures ANOVA?

variance between Iv levels / variance within IV levels

What are cell means?

mean scores for each condition

What are marginal means?

mean scores for single IV levels (ignoring other IV)

What are the three assumptions for two-way repeated measures ANOVA?

normality, sphericity (homogeneity of covariance), equivalent sample size

What is the parametric equivalent for factorial ANOVA?

there isn't one

What is the equation for partial eta squared?

SSM/ SSM + SSR

When are tests of simple effects needed?

when a significant interaction is found

What do tests for simple effects test?

compare across the levels of the main IV separately for each level of the secondary IV

What test is used for simple effects in a two way repeated measures ANOVA?

paired t tests (one for each level of secondary IV)

What does the test for simple effects that we use depend on in a two way mixed ANOVA?

if the main IV is between-subjects we do independent t test, if the main IV is within, we do paired t tests

What is the campaign against the abusive use of ANOVAs?

they are flexible, powerful technique appropriate for many experimental designs but good stats aren't a substitute for poo experimental designs, analyses should be considered at early stage

What do you have to ask before collecting any data?

do i have a clear research question? do i know what analyses i need? will i be able to carry out and interpret the results? have i considered potential cofounds? will i understand the answer i get?