Advances in Moderation Analysis: An Introduction to Mixture Regression

This webinar is available for free to NCFR members and to Certified Family Life Educators (CFLEs). Log in, or become an NCFR member or a CFLE, to access it.
April 30, 2019 11:00am - 12:00pm
CT
Free for NCFR members & CFLEs / $89 for nonmembers
Location
Online

W. Justin Dyer

Moderation analyses in research typically are conducted using either interaction terms or multiple group analyses. In this way, scholars can examine whether the relationship between variables differs across the levels of another variable.

However, it may be that the relationships among variables differ for certain “unobserved” or “latent” subgroups within a population (i.e., groups that are not directly observed). Mixture regression — a type of finite mixture model/latent class analysis — is a method to examine whether the relationships between independent and dependent variables differ for latent classes. That is, mixture regression identifies classes for whom the relationships between the independent and dependent variables differ. This method to examine moderation is highly flexible, identifying a multiple of groups for whom variables are differentially related and allows for examining the characteristics of those groups.

By attending this webinar — presented by W. Justin Dyer, Ph.D. — you’ll receive an accessible introduction to mixture regression and be able to understand:

  • basic concepts of moderation;
  • how mixture regression can address moderation questions that previous standard methods cannot; and
  • how to conduct a mixture regression.

You’ll learn how to identify subgroups of families and individuals who may not follow typical patterns, which can otherwise be difficult. You’ll learn about the theory and techniques related to mixture regression that will enable to you to employ this important new method. A "hands on" portion of the webinar will utilize MPlus and STATA software.

Approved for 1 CFLE contact hour of continuing education credit.

Intended Audience: Researchers, new professionals, students

Purchase On-Demand Recording
 


About the Presenter

W. Justin Dyer, Ph.D., is an associate professor at Brigham Young University, where he has taught courses on family and structural equation modeling. His research area includes fatherhood with a particular emphasis on fathers of children with disabilities and incarcerated fathers. His work also has focused on statistical innovations that enable researchers to better address family research questions, an area in which he has published several articles. He has taught numerous methodology workshops, including a yearly workshop on longitudinal structural equation modeling.

On-Demand Webinar Recording and Classroom Use

Even if you can't watch this webinar live, your registration will still grant you access to watch the recording at your convenience.

Classroom and departmental use licenses allow faculty members to share the video in class or embed the video in their online learning management system. Departmental use licenses allow more than on faculty to use the webinar in their class. We request that links or downloads are not shared with students.

License for classroom use by one professor is available for $129 for NCFR members, $199 for nonmembers.

License for departmental use (multiple professors) is available for $179 NCFR member / $319 nonmember.

Departmental license for CFLE-approved programs is $149.

Purchase a Webinar for Classroom Use >>