Transcript
COMPLEX EXPERIMENTAL DESIGNS
© 2009 The McGraw-Hill Companies, Inc.
FACTORIAL DESIGNS
Factorial Designs: Designs with more than one independent variable (or factor)
Simplest Factorial Design 2 x 2 factorial design
Has two independent variables Each independent variable has two levels
© 2009 The McGraw-Hill Companies, Inc.
Sample study: treating spider phobias
Therapy
Exposure only Exposure + Cognitive
Type of spider means:
Type of spiderReal spider 6 9 7.5
Fake spider 4 4 4
Therapy means:
5 6.5
© 2009 The McGraw-Hill Companies, Inc.
2 X 2: type of therapy by type of spider4 conditionsDV: score of spider phobia scale – higher the less afraid
FACTORIAL DESIGNS
Interpretation of Factorial Designs Main effects of an independent
variable when the levels of one IV differ while
ignoring the values of the other IV Interaction between the
independent variables when the effect of one IV depends on
the level of the other IV
© 2009 The McGraw-Hill Companies, Inc.
FACTORIAL DESIGNS
Interpretation of Factorial Designs Main effects of an independent variable Interaction between the independent variables
© 2009 The McGraw-Hill Companies, Inc.
Therapy
Exposure only
Exposure + Cognitive
Type of spider means:
Type of spider
Real spider 6 9 7.5
Fake spider 4 4 4
Therapy means:
5 6.5
Main effect of type of spider
Main effect of therapy
Interaction
Looking at the interaction: graph the cell means
© 2009 The McGraw-Hill Companies, Inc.
An interaction exists when lines are not parallel
© 2009 The McGraw-Hill Companies, Inc.
Is there an interaction?
© 2009 The McGraw-Hill Companies, Inc.
Notes on Graphing
Plot only cell means Use bar graphs
when the IV is categorical
Use line graphs when the IV is quantitative
© 2009 The McGraw-Hill Companies, Inc.
FACTORIAL DESIGNS
Factorial Designs with Manipulated and Non-manipulated Variables IV x PV designs
(Independent Variable by Participant Variable)
© 2009 The McGraw-Hill Companies, Inc.
Therapy
Exposure only
Exposure + Cognitive
Gender means:
genderMen 6 9 7.5
Women 4 4 4
Therapy means:
5 6.5
FACTORIAL DESIGNS
Outcomes of a 2 x 2 Factorial Design1. There may be a significant main effect for independent variable
A2. There may be a significant main effect for independent variable
B3. There may be a significant interaction between the independent
variables(See results in following slides)
© 2009 The McGraw-Hill Companies, Inc.
FACTORIAL DESIGNS
© 2009 The McGraw-Hill Companies, Inc.
FACTORIAL DESIGNS
© 2009 The McGraw-Hill Companies, Inc.
FACTORIAL DESIGNS
© 2009 The McGraw-Hill Companies, Inc.
FACTORIAL DESIGNS
© 2009 The McGraw-Hill Companies, Inc.
FACTORIAL DESIGNS
© 2009 The McGraw-Hill Companies, Inc.
FACTORIAL DESIGNS: 2 X 3
© 2009 The McGraw-Hill Companies, Inc.
Simple main effect of type of therapy at each type of spider
© 2009 The McGraw-Hill Companies, Inc.
Therapy
Exposure only Exposure + Cognitive
Type of spider means:
Type of spiderReal spider 6 9 7.5
Fake spider 4 4 4
Therapy means:
5 6.5
Simple main effect of type of spider at each type of therapy
© 2009 The McGraw-Hill Companies, Inc.
Therapy
Exposure only Exposure + Cognitive
Type of spider means:
Type of spiderReal spider 6 9 7.5
Fake spider 4 4 4
Therapy means:
5 6.5
FACTORIAL DESIGNS
Interactions and Simple Main Effects Simple main effect of type of therapy at each type of spider
© 2009 The McGraw-Hill Companies, Inc.
FACTORIAL DESIGNS
Interactions and Simple Main Effects Simple main effect of type of therapy at each type of spider Simple main effect of type of spider at each type of therapy
© 2009 The McGraw-Hill Companies, Inc.
vs
vs
FACTORIAL DESIGNS
Assignment Procedures and Factorial Designs Independent groups design
Participants are only in one level of an IV Repeated measures design
Participants are in all levels of an IV Mixed factorial design using combined
assignment One IV is independent, the other is repeated
© 2009 The McGraw-Hill Companies, Inc.
FACTORIAL DESIGNS
© 2009 The McGraw-Hill Companies, Inc.
top related