ABSTRACT
This chapter deals with between groups multi-factorial ANOVA, where more than one independent variable is manipulated or observed. When the authors analyse results in a two-way design they look at the main effect for factor A, the main effect for factor B and the interaction effect. If it then turned out that the 5 hours sleep groups were significantly superior in skill to the no sleep group overall then the interpretation of a main effect as showing superiority for 5 hours sleep would be misleading. A full calculation of effect sizes and power in this two-way design is provided on the Instructor and Student Resource website. In this three-way design there would be three main effects for the three independent variables, three two-way interactions and an overall three-way interaction. The effect sizes (partial eta squared) and 'observed power' are given in the right-hand columns.
