Experimental Designs Multiple-Group Designs Multiple-IV Designs

1
Experimental DesignsMultiple-Group
DesignsMultiple-IV Designs
2
Single IV Designs

• The multiple-group design

Independent Variable
Condition 2
Condition 3
Condition 1
3
Single IV Designs

• Bransford Johnson (1972)

4
Single IV Designs

• Bransford Johnson (1972)

Context
No context 1 rep
No context 2 reps
Context before
Context after
Partial context
5
(No Transcript)
6
Single IV Designs

• How to analyze a multiple-group design?

Context
No context 1 rep
No context 2 reps
Context before
Context after
Partial context
7
Single IV Designs

• How to analyze a multiple-group design?
• Analysis Oneway ANOVA

8
Single IV Designs

• Analysis Oneway ANOVA
• Null hypothesis?
• Alternative hypothesis?

9
Single IV Designs

• The multiple-group design Analysis

Independent Variable
Condition 2
Condition 3
Condition 1
10
Single IV Designs

• The multiple-group design Analysis

Independent Variable
Condition 2
Condition 3
Condition 1
Post-hoc comparisons
11
Single IV Designs

• The multiple-group design Analysis

Independent Variable
Condition 2
Condition 3
Condition 1
Post-hoc comparisons
12
Single IV Designs

• Bransford Johnson (1972)

Context
No context 1 rep
No context 2 reps
Context before
Context after
Partial context
3.6 (.64)
3.8 (.79)
8.0 (.65)
3.6 (.75)
4.0 (.60)
13
Single IV Designs
14
Single IV Designs

• Advantages of multiple-groups designs
• Can discover nonlinear relationships
• Test (and possibly rule out) potential
  alternative explanations

15
Multiple IV designs

• Factorial Designs
• Signal enhancers
• Advantages over single-IV designs
• Allow us to look at combinations of IVs at the
  same time
• More time efficient than running multiple
  single-IV studies

16
Factor A (First IV)
Level A2
Level A1
A1B1
A2B1
Conditions The combination of levels that
participants experience
Level B1
Factor B (Second IV)
A1B2
A2B2
Level B2
Levels Subdivisions of factors
Factors Major independent variables
17
Factorial Designs

• 2 x 2 design
• Number of numbers tells us how many factors
• Value of numbers tells us how many levels in each
  IV

18
Presentation Rate
2 sec
4 sec
Imagery
Training Type
Rote
19
Presentation Rate
2 sec
4 sec
Imagery
Training Type
Rote
20
Factorial Designs

• Main Effect
• Sole effect of one IV

21
Presentation Rate
2 sec
4 sec
17
23
20
Imagery
Training Type
12
15
18
Rote
14.5
20.5
22

• Main effect

23
Factorial Designs

• Main Effect
• Sole effect of one IV
• Interaction
• Joint, simultaneous effect of both IVs at the
  same time
• If you have a significant interaction, the
  effects of one IV depend upon the level of the
  other IV

24
Grant et al. (1998)
Study Environment
Silent
Noisy
14.3
12.7
12.8
Silent
Testing Environment
12.7
13.5
14.3
Noisy
12.8
13.5
25
(No Transcript)
26
Assigning Participants to Groups in Factorial
Designs

• Independent groups
• Randomly assigned to groups
• Correlated groups
• Matched pairs
• Repeated measures
• Natural pairs
• Mixed groups
• At least one IV has participants randomly
  assigned to groups, at least one IV has
  participants in correlated groups

27
Statistics in Factorial Designs

• Factorial ANOVA (in SPSS General Linear Model,
  Univariate Analysis of Variance)

28
Statistics in Factorial Designs

• Factorial ANOVA

29
Statistics in Factorial Designs

• Factorial ANOVA

30
Statistics in Factorial Designs

• Steps for interpretation (2 x 2)
• Is there a significant interaction?
• If yes, you must describe whats going on in the
  experiment using both IVs

31
(No Transcript)
32
Statistics in Factorial Designs

• Steps for interpretation (2 x 2)
• Is there a significant interaction?
• If yes, you must describe whats going on in the
  experiment using both IVs
• If no, ask
• Are there significant main effects?
• Interpret effect of factor A
• Interpret effect of factor B

33
Statistics in Factorial Designs

• How about a 2 x 3 factorial design?
• Is there a significant interaction?
• If yes, you must describe whats going on in the
  experiment using both IVs
• If no, ask
• Are there significant main effects?
• Significant effect of factor A?
• Significant effect of factor B?
• If significant, then what?

34
Statistics in Factorial Designs

• How many interactions in a 2 x 2 x 2 factorial
  design?