Experiments with more than one independent variable

Factor = Independent Variable

Two factors (IVs) create a two-way design
 

Design Names

Derived from number of variables and number of levels per variable

2 IVs with 2 levels each is a 2 X 2

3 IVs with 2 levels each is a 2 X 2 X 2

2 IVs, one with 2 levels and one with 3 levels is a 2 x 3
 

More Terminology

Each condition is also referred to as a cell

In a 2 X 2, there are 4 cells

In a 2 X 2 X 2 there are 8 cells
 

Main Effects

Effect of IV #1 on DV ignoring IV #2

Also effect of IV #2 on DV ignoring IV #1

How can we find main effects?
    - ANOVA source table
    - Line Graphs
 

Interactions

IV #1 affects DV differently depending on the level of IV #2

You must always explain your interactions because there are many kinds

Non-parallel line graphs suggest interaction
 

Finally...

A significant interaction makes it difficult to interpret main effects

Always focus on the higher order comparison, in this case the interaction

Crossover interaction is the ideal result

 
Back to Overheads Main Page
Back to 381-002 Main Page