Experimental Designs

1
Experimental Designs
2
Classifying Experimental Designs

• Observations in a study can be divided into two
  components
• Signal The key variablethe construct youre
  trying to measure
• Noise All random factors in the situation that
  make it harder to see the signal

3
Classifying Experimental Designs
4
Classifying Experimental Designs

• You want the signal to be high relative to the
  noise

5
Classifying Experimental Designs

• Classify experimental designs into to categories
• Signal Enhancers
• Factorial designs
• Noise Reducers
• Correlated groups designs
• Matched pairs, repeated measures,
  naturally-occurring pairs

6
Single IV Designs

• The basic two-group design

Independent Variable Customer Hearing
Experimental group
Control group
Customers were deaf
Customers were hearing
7
Single IV Designs

• Independent groups
• Randomly assigned to groups
• Correlated groups
• Matched pairs
• Repeated measures
• Natural pairs

8
Single IV Designs

• Advantages of independent-groups designs
• Simplicity
• In some contexts, it is impossible to use
  correlated groups
• Advantages of correlated-groups designs
• Controlwe have greater certainty of equality
• Statistical benefits

9
Statistical benefits

• Two sources of variability in your data
• The IV, or between-groups variability
• Error variability, or within-groups variability

10
Single IV Designs

• The multiple-group design

Independent Variable Type of Noise
Experimental group 2
Experimental group 3
Experimental group 1
White noise
Music
No noise
11
Single IV Designs

• The multiple-group design
• Analysis Oneway ANOVA

12
Single IV Designs

• The multiple-group design Analysis

Independent Variable Type of Noise
Experimental group 2
Experimental group 3
Experimental group 1
White noise
Music
No noise
13
Single IV Designs

• The multiple-group design Analysis

Independent Variable Type of Noise
Experimental group 2
Experimental group 3
Experimental group 1
White noise
Music
No noise
Post-hoc comparisons
14
Single IV Designs

• The multiple-group design Analysis

Independent Variable Type of Noise
Experimental group 2
Experimental group 3
Experimental group 1
White noise
Music
No noise
Post-hoc comparisons
15
Factorial Designs

• Multiple IV designs
• Signal enhancers

16
Factor A (First IV)
Level A2
Level A1
A1B1
A2B1
Level B1
Factor B (Second IV)
A1B2
A2B2
Level B2
Levels Subdivisions of factors
Factors Major independent variables
17
Analysis, example 1

• Main effect of time no effectof setting

18
Analysis, example 1

• Main effect of time

19
Analysis, example 1

• Main effect of setting

20
Analysis, example 2

• Main effect of setting,no effectof time

21
Analysis, example 2

• Main effect of time

22
Analysis, example 2

• Main effect of setting

23
Analysis, example 3

• Main effect of bothtime and setting

24
Analysis, example 3

• Main effect of time

25
Analysis, example 3

• Main effect of setting

26
Analysis, example 4

• Main effect of bothtime and setting, and
  aninteraction

27
Analysis, example 4

• Basicinteraction

28
Analysis, example 5

• Main effect of bothtime and setting, and
  aninteraction

29
Analysis, example 5

• Crossover interaction