1. Between-Subjects Factorial Design

2. Chapter Objectives
Learn how to test more than one independent variable in the same experiment
Learn about the main effects and interactions between variables
Learn how to diagram and label factorial experiments
Understand how to interpret effect from factorial experiments

3. More than one Independent Variable
In real life, variable rarely occur alone.
Experiment with more than one independent variables are efficient and provide more information than experiments with one independent variable.
Factorial designs – designs with two or more independent variables
Independent variables are called factors
Two factor experiment – the simplest factorial design

4. Factorial Designs
They give us information about the effects of each independent variable in the experiment – main effects
They enable us to answer the question: How does the influence of one independent variable affect the influence of the other variable.

5. The Main Effects
The action of a single independent variable in an experiment.
A change in behavior associated with a change in the value of a single independent variable or factor in the experiment.
There are as many main effects as there are factors

6. Looking for Interactions
Factorial design allows us to test for the relationships between the effects of different independent variables
An interaction is present if the effect of one independent variable changes across the levels of another independent variable.
Alcohol alone or sleeping pill alone can reduce anxiety, together, they can be fatal
The effects of one factor will change depending of the level of the other

7. Looking for Interactions

8. Looking for Interactions

9. Looking for Interactions

10. Looking of Interactions
An interaction can tell us that there are limits to the effect of one or more factors
Two independent variable = one interaction
More than two independent variables result to more complex interactions – higher-order interactions
It is possible to have interactions but no main effects
Or significant effect with no interaction

11. Laying Out a Factorial Design
Step 1: Indicate the 2 independent variables
Factor 1 (Type of Name)
Factor 2
(Lenght of Name)

12. Laying Out a Factorial Design
Step 2: Indicate the levels of Factor 1
Factor 1 (Type of Name)
Given Name
Nickname
Factor 2
(Lenght of Name)

13. Laying Out a Factorial Design
Step 3: Indicate the levels of Factor 2
Factor 1 (Type of Name)
Given Name
Nickname
Short
Factor 2
(Lenght of Name)
Long

14. Laying Out a Factorial Design
Step 4: Indicate the 4 treatment conditions
Factor 1 (Type of Name)
Given Name
Nickname
Short
Factor 2
(Lenght of Name)
Long

15. Laying Out a Factorial Design
Step 4: Indicate the 4 treatment conditions
Factor 1 (Type of Name)
Given Name
Nickname
Short given name
Short
Factor 2
(Lenght of Name)
Long

16. Laying Out a Factorial Design
Step 4: Indicate the 4 treatment conditions
Factor 1 (Type of Name)
Given Name
Nickname
Short given name
Short nickname
Short
Factor 2
(Lenght of Name)
Long

17. Laying Out a Factorial Design
Step 4: Indicate the 4 treatment conditions
Factor 1 (Type of Name)
Given Name
Nickname
Short given name
Short nickname
Long give name
Short
Factor 2
(Lenght of Name)
Long

18. Laying Out a Factorial Design
Step 4: Indicate the 4 treatment conditions
Factor 1 (Type of Name)
Given Name
Nickname
Short given name
Short nickname
Long give name
Long nickname
Short
Factor 2
(Lenght of Name)
Long

19. Describing the Design 2 x 2 (two by two) factorial design
Other methods of describing design variables:
Factor-Labeling Method
2 x 2 (Type of Name x Length of Name) between-subjects factorial design
2 (name type) x 2 (name length) between-subjects factorial design
Factor and Levels Method
2 x 2 (Type of Name: given, nickname x Length of Name: short, long) between-subjects factorial design
2 (given name or nick name) x 2 (short or long name) between-subjects factorial design

20. Describing the Design 2 x 3 x 2 factorial design
There are 3 numbers which means there are 3 factors or independent variables
Factor 1 has two levels
Factor 2 has three levels
Factor 3 has two levels
There are 12 separate conditions

21. Describing the Design Factor 1: Sex of Communicator Male Communicator
Female Communicator
Factor 2:
Communicator attractiveness
Low
Moderate
High
Factor 3: Subject sex
Male
Female
There are 3 independent variables: Sex of the person delivering the persuasive communication, attractiveness of the communicator, and the sex of the subject listening to the communication. It is hypothesized that subjects will be more persuaded by a message delivered by a more attractive communicator, but only when the communicator is the opposite sex from the subject

22. Sample Experiment
Women will eat less in the presence of an opposite-sex partner than in the presence of another of the same sex. Men’s eating behavior will not be influenced by their partners gender.
Factor 1 (Sex of the Subject)
Male
Female
Male
Factor 2:
(Sex to the subject’s Partner)
Female
DV: Eating behavior

23. Sample Experiment
Significant effect for subject sex; significant interaction for females in the presence of males

24. Sample Experiment Findings:
In general, men ate more crackers than women
In general, the gender of the partner did not alter the number of crackers eaten
There was a significant interaction between the sex of the subject and the type of partner.

25. Sample Experiment: Hypothetical Results
No significant main effect or interaction

26. Sample Experiment: Hypothetical Results
Significant main effect for partner’s sex

27. Sample Experiment: Hypothetical Results
Significant main effect for subject sex

28. Sample Experiment: Hypothetical Results
Two Significant main effects

29. Sample Experiment: Hypothetical Results
Significant Subject sex by Partner sex interaction ; eating depends on both factors

30. Sample Experiment: Hypothetical Results
Maximum significant subject sex by partner sex interaction; crossover interaction

31. Sample Experiment: Hypothetical Results
Significant subject sex main effect. Significant partner sex main effect, and a significant subject sex by partner sex interaction

33. Measuring interactions in one key reason for doing factorial research.
By looking at two or more variables together, we can assess whether the effect of one variable depends on the level of another variable

34. Choosing a Between Subjects Design
Practical reasons for keeping factorial designs simple:
More treatment condition means more subjects
More treatment condition means more time to run the experiment
More treatment condition means more time to do the statistical analysis
Complicated design are virtually uninterpretable
Four way interactions are practically impossible to conceptualize and explain
2 x 2 factorial design has 3 possible effects
2 x 2 x 2 factorial design has 7 possible effects

35. Will I need more than two levels of my independent variable to test
What is my hypothesis?
Does my hypothesis have more than one independent variable?
YES NO
I need a factorial design
Will I need more than two levels
of my independent variable to test
my hypothesis?
Now I will diagram it.
Do I know of a subject variable
that will have a big effect on
my dependent variable?
I need a multiple-group design
Can I measure this variable?
I will use the two-independent-
groups design and assume
that randomization controls
for confounding
from any subject variables
i will use
the two-matched-groups
design to be sure
my groups start
out the same on
this variable
Forget it. I will
have to use the two-
independent=groups
design and assume
that randomization
takes care of this
variable
How many
treatment conditions
do I have?
Is one of them a
control condition?
Good
Why not?