Two-way between-subjects ANOVA Flashcards
factorial anova
more than one factor
variables can interact and the effects of one. on another may differ according to te level of the other factor
what does 2-way b-s anova allow?
comparisons of 2 or more groups at the same time but groups can have more than one IV
2+ IV with 2+ levels
look at the main effect of each IV alone and the main effect of the interaction
SPSS outputs for 2-way between-subjects anova
- Between subjects factors (levels of factors)
- Descriptive statistics (means and IDs for the factors separately and their interaction)
- Levene’s test of between-subjects effects (‘based on mean’ row - p>.05 (non sig) shows variances within-groups aren’t significantly different from each other)
- Tests of between-subjects effects (ANOVA stats for reporting main effect/interaction) (F equation with eta square)
- Profile plots (graph of interaction where parallel lines show no significant interaction)
equation for effect size for 2-way b-s anova (eta square)
sum of squares for the factor or interaction / total sum of squares
Cohen’s guidelines for effect sizes
s = 0.01
m = 0.059
l = 0.138
formally reporting anova
- ANOVA type, effect of factor/IV on DV
- Present means and SDs
- Mention assumptions
- Report ANOVA results giving DF, F ratio and p value
- Report effect sizes and what it means
- Report all main effects and interactions
- Report comparisons
- Interpret in words
Interpreting factorial anova
- ANOVA table shows which main effects and interaction terms are significant
- main effects = follow up with (un)planned comparisons for factors wth more than 2 levels
- Interactions = follow up with simple effects (using t-tests)
What is the Bonferroni adjustment/correction and why use it?
used to reduce type 1 error
p<.05/no. of comparisons you are making