Eastern Communication Association

Integrating Mediation and Moderation Analysis:
Fundamentals using PROCESS

A Half-Day Workshop by Andrew F. Hayes
Thursday, April 2, 1-5 PM

Conditional process analysis, also known as the analysis of moderated mediation, is the integration of mediation and moderation analysis and used when one’s analytical goal is to describe and understand the conditional nature of the mechanism or mechanisms by which a variable transmits its effect on another (see Hayes, 2018, Introduction to Mediation, Moderation, and Conditional Process Analysis). After overviewing the fundamentals of mediation and moderation analysis, this workshop introduces the fundamentals of conditional process analysis and its implementation using the PROCESS tool for SPSS or SAS.   Using OLS regression-based path analysis, it covers the estimation of various classes of models which allow indirect and/or direct effects to be moderated, the estimation of conditional indirect effects, testing a moderated mediation hypothesis, and how to compare conditional indirect effects.    

This workshop will be useful for those who want to learn how to conduct mediation, moderation, and conditional process analysis using SPSS and SAS. Participants should have a basic working knowledge of the principles and practice of multiple regression and elementary statistical inference. No knowledge of matrix algebra is required or assumed.

Because this is a hands-on workshop, participants are strongly encouraged to bring their own laptops (Mac or Windows) with a recent version of SPSS Statistics (version 23 or later) or SAS (release 9.2 or later) installed.  SAS users should ensure that the IML product is part of the installation.  Files needed for the workshop will be distributed by USB at the beginning of class.  Because the room will probably not have many power outlets available, participants should make sure they have sufficient battery charge for 4 hours of continuous use. 

About the Instructor

Andrew F. Hayes is Professor of Psychology and Professor of Communication at The Ohio State University and a Visiting Scholar at the Stephen J. Smith School of Business at Queen’s University, Kingston, Ontario, Canada. He is also an instructor for Statistical Horizons LLC and the Global School in Empirical Research Methods. His research focuses on linear models and their application to modeling the mechanisms and contingencies of effects. His methodological work has appeared in Psychological Methods, Multivariate Behavioral Research, Behavior Research Methods, Behaviour Research and Therapy, Psychological Science, Australiasian Marketing Journal, American Behavioral Scientist, Communication Methods and Measures, Human Communication Research, Journal of Communication, and Communication Monographs, among others. He is the author of three books: Regression Analysis and Linear Models (2017, The Guilford Press), An Introduction to Mediation, Moderation, and Conditional Process Analysis (2018, The Guilford Press), and Statistical Methods for Communication Science (2005, Routledge).  According to Google Scholar, his work has been cited over 90,000 times by scholars throughout the world in nearly every empirical research area. He can be found in cyberspace at www.afhayes.com

Return to Top of Page

Mediation Analysis with Distinguishable Dyadic Data

A Half-Day Workshop by Jacob J. Coutts, Tao Jiang, and Andrew F. Hayes
Friday, April 3, 8 AM - Noon

When data are collected from distinguishable dyad members (e.g., husband/wife, boss/employee, parent/child), mediation analysis can be used to understand the mechanism(s) by which causal effects operate within and between dyad members. In this workshop, after addressing some of the shortcomings of structural equation model approaches and the PROCESS macro for mediation analysis with distinguishable dyadic data, we illustrate comparable analyses using the easy-to-use MEDYAD macro for SPSS and SAS. We discuss various dyadic mediation models and demonstrate how they can be easily estimated using MEDYAD. Some of the useful features of MEDYAD are discussed, including the ability to easily compare indirect effects within and between dyad members in order to better understand the complex dynamics of mutual causal influence in dyadic relationships.

This workshop will be useful for those who want to learn how to conduct mediation analysis with distinguishable dyadic data. Participants should have a basic working knowledge of the principles and practice of multiple regression and elementary statistical inference. No knowledge of matrix algebra is required or assumed.

Because this is a hands-on workshop, participants are strongly encouraged to bring their own laptops (Mac or Windows) with a recent version of SPSS Statistics (version 23 or later) or SAS (release 9.2 or later) installed. SAS users should ensure that the IML product is part of the installation. Files needed for the workshop will be distributed by USB at the beginning of class. Because the room will probably not have many power outlets available, participants should make sure they have sufficient battery charge for 4 hours of continuous use.    

About the Instructors

Jacob J. Coutts is a second-year graduate student pursuing a Ph.D. in Quantitative Psychology and a Master of Applied Statistics under the advisement of Andrew F. Hayes. His research presently focuses on linear models and their application to modeling mechanisms of effects, especially as they pertain to distinguishable dyads. He also does research investigating methods of comparing these mechanisms (i.e., indirect effects) with each other. With his training in numerous programming languages (e.g., R, GAUSS, SPSS), he strives to make advancements in modeling techniques more accessible to other researchers by developing user-friendly software. For more information, visit his website at www.jjcoutts.com.

Tao Jiang is a fifth-year Ph.D. candidate in social psychology at The Ohio State University. His research focuses on how people’s motivations and beliefs influence their own and their partners’ interpersonal behaviors in close relationships and the intrapersonal and interpersonal mechanisms through which they operate. His research work has appeared in Self and Identity, Scandinavian Journal of Psychology, Asian Journal of Social Psychology, among others. Moreover, he has obtained a lot of training and experience in advanced statistical skills. Especially, he has been working extensively on analyzing dyadic and longitudinal data using multilevel modeling and structural equation modeling.    

Andrew F. Hayes is Professor of Psychology and Professor of Communication at The Ohio State University and a Visiting Scholar at the Stephen J. Smith School of Business at Queen’s University, Kingston, Ontario, Canada. He is also an instructor for Statistical Horizons LLC and the Global School in Empirical Research Methods. His research focuses on linear models and their application to modeling the mechanisms and contingencies of effects. His methodological work has appeared in Psychological Methods, Multivariate Behavioral Research, Behavior Research Methods, Behaviour Research and Therapy, Psychological Science, Australiasian Marketing Journal, American Behavioral Scientist, Communication Methods and Measures, Human Communication Research, Journal of Communication, and Communication Monographs, among others. He is the author of three books: Regression Analysis and Linear Models (2017, The Guilford Press), An Introduction to Mediation, Moderation, and Conditional Process Analysis (2018, The Guilford Press), and Statistical Methods for Communication Science (2005, Routledge).  According to Google Scholar, his work has been cited over 90,000 times by scholars throughout the world in nearly every empirical research area. He can be found in cyberspace at www.afhayes.com

Return to Top of Page

Qualitative Research Methods and Practice:  A Hands-On Workshop

Heather Stuckey, D. Ed., The Penn State Milton S. Hershey College of Medicine
Friday, April 3 1 – 5 PM

This experience is designed for both new qualitative researchers, and those who would like to build upon their qualitative research and perspectives or explore the potential of mixed methods approaches to research design. The first part of the workshop will offer participants hands-on practice in research and interview question development to make a solid foundation for a qualitative study.   This will be followed by discussion of well-developed qualitative studies.  The second part of the afternoon will involve the development of codes and themes.  We will conclude with work group session to practice coding, interpretation, and development of themes from codes, and feedback from the facilitator.    By the end of the session, participants will understand the benefits of qualitative research, and be able to develop interview guides and an initial codebook with interpretation of themes.  There will also be opportunity for lots of networking, sharing of information and ideas!

About the Presenter

Heather Stuckey, D.Ed., is Associate Professor of Medicine, Humanities and Public Health Sciences at the Penn State University Hershey College of Medicine.  Dr. Stuckey’s expertise is in qualitative research and mixed methods, generally leading to interventions to improve diabetes management and psychosocial outcomes. She was lead qualitative investigator for the DAWN2 (Diabetes Attitudes, Wishes and Needs) study to identify psychosocial needs of people with diabetes, health care providers and family members around the world.  Dr. Stuckey is extending that research to the UK, where she is studying the interaction of severe mental illness (SMI) with chronic disease.  She currently holds an NIDDK DP3 grant to analyze patient blog use and provider perceptions to identify barriers and facilitators to self-management. 

She works on other funded opportunities, all related to researching quality of life and pathways to improvement in living with chronic disease.  Dr. Stuckey has published multiple articles on diabetes and psychosocial factors, social networking support, habits of successful diabetes/weight loss practices, and the use of creative expression in diabetes.  She is also Director of Research for the Foundation for Art & Healing, and has a strong research focus on determining strategies for incorporating the arts into the patient experience.

Return to Top of Page