Latent Profile Analysis

July 21-22, 2025, during Rocky Mountain Methodology Academy 2025

$950 (Canadian dollars) + 5% GST

Register through the RMMA 2025 page

Many researchers are interested in studying and addressing questions focused on group differences. Do students in two different schools have different academic performance? Do patients receiving a new treatment have better outcomes than patients receiving a traditional treatment?  But… what if the groups a researcher wanted to study cannot be directly observed? Subgroups that cannot be directly observed can involve those that reflect work teams that have distinct configurations of interpersonal conflict, or employees that have unique combinations of organizational commitment mindsets, or individuals that demonstrate discrete types of coping mechanisms, among many others. Latent profile analysis (LPA) is a method that can be used to identify heterogenous subgroups and address important research questions when group membership cannot be directly observed (i.e., is latent). 

LPA is a member of the mixture modeling family of statistical models that assume an observed sample of data includes a mixture of distinct distributions, each representing a latent subgroup. LPA, and other mixture models, are becoming increasingly popular in areas of psychology, marketing, education, management, and many other areas of the social sciences, as a powerful tool for identifying unique, unobserved subgroups. 

Although LPA is similar to traditional cluster analyses, LPA leverages a more advanced and flexible modeling framework, providing an intuitive and model-based approach to classify individuals, teams, organizations, or other units of analysis. This flexibility provides an opportunity to explore predictors and outcomes of subgroup membership, as well as more complex analytical components involving mediation, moderation, and conditional effects that cannot be done with traditional clustering methods.

This course will introduce participants to LPA, focusing on applications in the social, educational, health, and management sciences. This course will provide participants with the theoretical and conceptual background and applied analytical skills needed to specify an appropriate model, interpret the results, and thoroughly address research questions using LPA. Using both R/RStudio and Mplus for different examples, this course offers a non-technical introduction to LPA and is paired with extensive worked examples and hands-on activities, as well as ample opportunities to discuss participants’ own LPA-based research ideas.

INSTRUCTOR: Matthew McLarnon PhD

In this course, participants will be introduced to a wide-range of topics, including:

  • The underlying principles of LPA and other forms of mixture models
  • Conceptual background on LPA and other mixture models such as latent class analysis (LCA) and growth mixture model (GMM)
  • Brief overview and background on both R and RStudio, as well as the Mplus software environments
  • Steps involved in specifying an appropriate analytical model, model fitting, model selection, and also diagnosing and troubleshooting common issues
  • Including and interpreting predictors and outcomes
  • Introducing advanced LPA models that contain mediation, moderation, and/or conditional effects

This course meets in person usually over two days from 9am to 5pm each day.

The course is designed to be both comprehensive and practical. Thus, this course will first introduce LPA within R using the TidySEM package. Subsequently, we will explore how LPA has been implemented within Mplus, given its flexibility and advanced features. 

Participants will engage in hands-on exercises using R and Mplus, ensuring they can apply what they learn directly to their own research projects. All Mplus examples used in the course will be compatible with the free Demo Version available from https://statmodel.com/demo.shtml. The full version of Mplus is not required.

Because this is a hands-on course, learners are encouraged to bring a laptop to class with up-to-date versions of R/RStudio and Mplus (again, the Demo version will be fully usable for our in-class examples). All data files and syntax will be available to participants so that all material can be downloaded onto personal laptops.

 

This course will be helpful for researchers in numerous fields – including psychology, sociology, education, business, human development, social work, public health, communication, etc. – who want to understand how to leverage LPA and other mixture models to explore their substantive research questions. Although this course will include a primer and background on the R and RStudio environments, as well as the Mplus software, participants should be comfortable conducting analyses covered by a typical introductory-level graduate-level statistics course (i.e., familiar with ANOVA and multiple regression). Previous experience with R/Rstudio or Mplus is not required.

Upon completing this course, participants will

  • understand the conceptual and theoretical background on LPA, and other mixture models
  • know how to make informed choices when specifying, fitting, evaluating, interpreting, and troubleshooting LPA models
  • be able to implement modeling extensions to include predictors and outcomes in an LPA model and interpret the results
  • understand the basic ideas underpinning extensions to advanced applications of LPA models which can include mediation, moderation, and/or conditional effects
  • be able to implement LPA in R/RStudio and Mplus

A certificate of completion from the Canadian Centre for Research Analysis and Methods is provided at the end of the course.