Psychology 2800E Lecture Notes - Lecture 15: Factorial Experiment, Observational Error
Document Summary
A complex design is one in which more than one variable is manipulated at the same time. Each variable is manipulated at all levels of each other variable. The effect of one variable in a multi-variable design, ignoring all other variables this means the average of a1b1 and a1b2. Comparing the average of a1 and average of a2 gives us the main effect of a. This gives us no information about the effect of variable b. The effect of one variable in a multi-variable design, observed at one level of a second variable. [a1b1 - a2b1] gives the simple main effect of a at level b1 of variable b. [a1b2 - a2b2] gives the simple main effect of a at level b2 of variable b. [a1b1 - a1b2] gives the simple main effect of b at level a1 of variable a. [a2b1 - a2b2] gives the simple main effect of b at level a2 of variable a.