Chapter 12: The Nuts and Bolts of Multi-factor experiments
Factorial Design Factorial design: an experiment or quasi-experiment that includes more than one independent variable. McBride, The Process of Research in Psychology. Third Edition © 2016 SAGE Publications, Inc.
Factorial Design McBride, The Process of Research in Psychology. Third Edition © 2016 SAGE Publications, Inc.
Factorial Design Many experiments conducted by researchers are factorial designs, meaning they contain more than one independent variable. The primary advantages of a factorial design over simpler experiments are that a researcher can be more efficient in testing the effects of multiple independent variables in one experiment and can also examine the combined effects of those independent variables on the dependent variable. McBride, The Process of Research in Psychology. Third Edition © 2016 SAGE Publications, Inc.
Factorial Design Testing the combined effects of the independent variables is the unique feature of factorial designs. Without including multiple independent variables in a single experiment, we would not be able to detect the different effects a factor might have on behavior in different situations. McBride, The Process of Research in Psychology. Third Edition © 2016 SAGE Publications, Inc.
Factorial Design When the effect of one independent variable (e.g., treatment type) depends on the levels of another independent variable (e.g., short or long outing), this is called an interaction effect. Interaction effect: tests the effect of one independent variable at each level of another independent variable in an ANOVA. McBride, The Process of Research in Psychology. Third Edition © 2016 SAGE Publications, Inc.
Factorial Design In factorial designs, the comparison of the mean scores for the levels of one independent variable is the test of the main effect of that independent variable. The main effect is one type of effect tested in an analysis of variance (ANOVA). The other type of effect tested in an ANOVA is an interaction effect. McBride, The Process of Research in Psychology. Third Edition © 2016 SAGE Publications, Inc.
Factorial Design McBride, The Process of Research in Psychology. Third Edition © 2016 SAGE Publications, Inc.
Factorial Design McBride, The Process of Research in Psychology. Third Edition © 2016 SAGE Publications, Inc.
Factorial Design McBride, The Process of Research in Psychology. Third Edition © 2016 SAGE Publications, Inc.
Factorial Design McBride, The Process of Research in Psychology. Third Edition © 2016 SAGE Publications, Inc.
More About Interactions Graphs are useful in determining the type of interaction that occurred and should be followed up with additional statistical tests to determine exactly where the differences between the conditions in the interaction are in order to best describe the interaction effect. The follow-up statistical tests are called simple effects tests. McBride, The Process of Research in Psychology. Third Edition © 2016 SAGE Publications, Inc.
More About Interactions Interactions between independent variables can reveal interesting effects of the variables beyond what is seen in the main effects of each variable. In the results shown in Figure 12.5, there would be no main effects of the independent variables but there is a clear interaction of these variables. McBride, The Process of Research in Psychology. Third Edition © 2016 SAGE Publications, Inc.
More About Interactions McBride, The Process of Research in Psychology. Third Edition © 2016 SAGE Publications, Inc.
More About Interactions McBride, The Process of Research in Psychology. Third Edition © 2016 SAGE Publications, Inc.
Experiment Examples Cognitive Biological Social Developmental McBride, The Process of Research in Psychology. Third Edition © 2016 SAGE Publications, Inc.