• All ANOVA calculations require that we partition the total variance in a set of scores into specific components that represent factors in our design.
  
  
• The MS_{within groups} is used as the error term for all effects in a between-groups factorial ANOVA.
  
  
• If you obtain a negative value when calculating effects in ANOVA, you need to check your work. Effects are estimates of variance; there can be no negative variance estimates.
  
  
• F ratios in ANOVA all have the same form: MS_effect/MS_error.
  
  
• We can test both main and interaction effects in factorial ANOVA. If we obtain a significant interaction effect, we do not interpret any significant main effects.

Test your knowledge of this workshop by taking the Workshop Quiz.