Factorial Design

Factorial designs tests whether independent variables (factors) affect different kinds of people or whether it affects people in different situations in the same way Can be a way to test moderating variables Also a way to test whether an effect generalizes to other groups (external validity) Can also be used in theory testing since many theories generate specific hypotheses about how variables interact with each other.

Quantity of numbers

the number of IVs

Value of each number

how many levels there are for each IV

Product of the numbers

number of cells or how many conditions there are

participant variables

A variable whose levels are selected, not manipulated

2x2 design

IVB

IVA

Cell Means

mean DV of all subjects in each cells

mean of all the numbers here

Marginal Mean

mean DV of all subjects in one condition of an IV.

CBT, Pschodynamic, General Anxiety, Phobia

Main Effects

•when one independent variable has an overall effect on a dependent variable (ignoring the other IV!)

- Uses marginal means

Interaction Effect

•Refers to when the effect of one independent variable depends on another independent variable

•Looking at one level of the independent variable at both levels of the other independent variable

•Uses cell means

•“Differences in differences”

How to tell if an interaction exists

•Patterns of data that include interactions can be identified and described using the “it depends” approach.

•the simple effect of one IV is different at different levels of the other IV

Different Differences

Interactions based on graphs

•If the lines are parallel, there is probably not an interaction.

•If the lines are not parallel, there is probably an interaction.

types of effects by graph

types of effects by graph part 2

describing main effect

“There is a main effect for training type such that the vocab training group scored higher than the no training group.”

“There is no main effect for test type.”

When describing a main effect for training type, you do not mention test type (and vice versa)!

Describing interaction

•Every interaction is different—there is no single sentence structure.

•Try the “simple main effects” strategy.

“There is an interaction, such that for verbal tests, vocabulary training did better than no training. But for math tests, the two training groups scored the same. “