Chapter 12: Experiments with More Than One IV Flashcards
Interaction effect
Whether the effect of the original IV depends on the level of another IV (aka moderation). “Difference in differences”
Factorial design
A design in which there are two or more independent variables.
Cell
A condition in an experiment; in a factorial design, a cell represents one of the possible combinations of two IVs.
Participant variable
A variable whose levels are measured, not manipulated.
Main effect
The overall effect of one IV on the DV, averaging over the levels of the other IV.
Marginal means
The arithmetic means for each level of an IV, averaging over levels of the other IV.
What can a factorial design do?
- Test limits: Test whether an IV affects different kinds of people in different ways/situations (form of external validity). Does it affect everyone the same way?
- Test theories: Many theories make claims about how and why variables affect each other. Use a factorial design to test this theory. Why does it happen?
- You can examine potential interactions.
Crossover interaction
The lines of each IV cross over each other on the line graph.
Spreading interaction
The lines of each IV meet almost meet at one point and spread outwards in different directions.
Interpreting factorial results
If there is a significant interaction, then the interaction is the most important effect.
The main effects are not meaningful because they disregard the important influence of the other IV.
If no significant interaction, then the main effects are interpreted.
It’s easier to detect an interaction using a figure.
Are the lines parallel?
Not parallel= probably an interaction.
Parallel = probably no interaction.
Independent groups design
Both IVs are manipulated as independent (between) groups. Participants in each cell are different from each other.
Within-groups design
Both IVs are manipulated as within groups.
For a 2x2 design, there is only one group of participants and all participants get the conditions from all 4 cells.
Mixed
One IV is between groups, the other IV is within groups.
Increasing levels of IV
You can add more levels to your IV. Instead of small and large bottles, you could do small, medium and large bottles.
__ x ___ design: The quantity of numbers represents the number of IVS. The value of the numbers represents the levels of each IV.
When you multiple the numbers, you get the number of cells.
Example: 2 x 3
2 independent variables.
1st IV has 2 levels, 2nd IV has 3 levels.
Example: 2 x 2 x 2 design (called a three-way design).
Increases the number of differences to examine: 3 main effects, 3 two-way interactions, 1 three-way interaction.
Identifying interactions from a bar graph
Connect bars of same color to each other. If lines cross, there is an interaction. If lines are parallel/don’t cross, there is no interaction.
