Mediation, Moderation, and Conditional Process Analysis

Date/Time
Date(s) - 06/06/2016 - 06/07/2016
All Day

Categories No Categories

Start Date: 6/6/2016

End Date: 6/7/2016

Location:

Queensland University of Technology

Gardens Point Campus, 2 George Street

Brisbane, Australia

Website: http://statisticalhorizons.com/seminars/public-seminars/mediation-and-moderation-australia16

Contact: 610-642-1941

Taught by Andrew Hayes, Ph.D. 

This seminar focuses on two topics in causal analysis that are closely related and often confused. Suppose we have three variables, X, M and Y. We say that M is a mediator of the effect of X on Y if X carries its influence on Y at least partly by influencing M, which then influences Y. This is also known as an indirect effect of X on Y through M. On the other hand, we say that M moderates the effect of X on Y if that effect varies in size, sign, or strength as a function of M. This is also known as interaction.

Although these concepts are fairly simple, the statistical issues that arise in estimating and testing mediation and moderation effects turn out to be rather complex and subtle. Andrew Hayes has been among the leading recent contributors to the literature on these methods. He has developed powerful new methods for estimating mediation and moderation effects and special software tools that can be used with SAS or SPSS.   

In this seminar, you will learn about the underlying principles and the practical applications of these methods. The seminar is divided roughly into three parts:

1. Partitioning effects into direct and indirect components, and how to quantify and test hypotheses about indirect effects.

2. Estimating, testing, probing, and visualizing interactions in linear models.

3. Integrating moderation and mediation by discussing how to estimate conditional indirect effects, determine whether an indirect effect is moderated (moderated mediation) and whether moderated effects are mediated (mediated moderation).

Computer applications will focus on the use of OLS regression and computational modeling tools for SPSS and SAS (including the PROCESS add on developed by Hayes).  When appropriate, some Mplus code will be provided for those interested, but structural equation modeling and Mplus will not be the emphasis of this seminar.

Because this is a hands-on course, participants are strongly encouraged to bring their own laptops (Mac or Windows) with a recent version of SPSS Statistics (version 19 or later) or SAS (release 9.2 or later) installed. SPSS users should ensure their installed copy is patched to its latest release. SAS users should ensure that the IML product is part of the installation. You should have good familiarity with the basics of ordinary least squares regression (although an overview of OLS will be the first topic of the course), as well as the use of SPSS or SAS. You are also encouraged to bring your own data to apply what you’ve learned.