Mediation and moderation are two powerful statistical techniques that focus on understanding the mechanisms of effects. Mediation focuses on modeling the process by which effects influence outcomes whereas moderation focuses on whether an effect is stronger under certain conditions. Together, these techniques can illuminate interesting patterns in your data. Mediation and moderation can be easily implemented in SPSS via a macro developed by Andrew Hayes called PROCESS.
This seminar will be of interest to researchers in economics, education, medicine, psychology, and sociology. Some examples of research questions that can be answered with PROCESS:
• Is the effect of an intervention in at-risk couples on ratings of child abuse potential mediated by parental conflict about the child?
•Is the effect of students’ self-perception of ability on the level of engagement in school moderated by ethnic identity?
• Does a mediator exist in certain conditions of your experiment but not the others?
• Are there multiple mediators of your effect, and if so, are some mediators significantly stronger than others?
This methodology seminar will provide a brief introduction to mediation and moderation, an orientation to PROCESS in SPSS, and a worked example of a mediation and moderation analysis. We will also briefly highlight some of the more advanced analyses available through PROCESS.
Registration Info: REGISTER HERE
Tina Chen is a fifth-year doctoral student in the Cognitive Psychology division who has been consulting with the Center for Research on Families since summer 2014. She is trained in a variety of statistical techniques, including ANOVA, regression, hierarchical linear modeling, structural equation modeling, mediation, moderation, and Bayesian statistics.
Daniel Rovenpor is a Graduate Methodology Consultant and is a PhD candidate in Social Psychology. His research focuses on the role of emotion in cognitive, motivational, and political processes. He is trained in numerous statistical techniques, including, ANOVA, regression, conditional process modeling (moderation and mediation), hierarchical linear modeling, and structural equation modeling.