Chapter 13: Factorial ANOVA

factorial ANOVA

considers 2 or more factors - each factor with 2 or more groups/levels

2-factor ANOVA

can be independent variables and/or quasi-independent variables; analyze main effect for each factor and interaction between (compute a separate f-ratio for each)

quasi-independent variable

not manipulated (ex: metabolism)

main effects

effect of one factor, ignoring levels of other factor; ex: main effect of Factor A (diet) and main effect of Factor B (exercise)

interaction b/w factors

effect of one factor depends on the level of a second factor; ex: effects of therapy on depression scores depends on whether or not the patient is on medication

main effects hypothesis

mean differences among levels of one factor

interaction hypotheses

the effect of one factor depends on the level of another factor

ordinal interactions

occur when a factor seems to have more of an effect at one level of a second factor than another level) - spreading

disordinal interactions

indicate a factor has one kind of effect at one level and a different effect at another level - crossover

distinct F-tests

main effect of Factor A, main effect of Factor B, interaction of A and B

F

(variance b/w treatments)/(variance if there is no effect - within)

what happens if you have a sig interaction?

ignore any sig main effects

what happens if you have no sig interaction?

examine main effects