Multilevel Structural Equation Modeling

Multilevel structural equation modeling (MSEM) is the most general and flexible instantiation of the general linear model. It breaks down artificial barriers between model types and allows investigators in the health and behavioral science maximal ability to pose sophisticated and complex questions. This course will sharpen your thinking and offer new conceptual insights regardless of the data type that is typical in your work. Using MSEM in Mplus also offers a wealth of advantageous tools, even for simple models.

Instructor:
Aidan Wright, PhD (University of Michigan)

Workshop Dates and Times:
Recorded on July 10-12, 2024
Now available for self-paced learning

Workshop Format:
Three-Day Online Workshop Recording

When researchers collect repeated measures within-person, as is the case in ambulatory assessment and other intensive longitudinal designs, this creates nested or hierarchical data structures. Typically, this requires multilevel or mixed effects models (MLM) to analyze correctly. At the same time, standard multilevel models are limited because they are not fully multivariate and can only include observed predictors. To address these limitations, researchers can use Multilevel Structural Equation Modeling (MSEM).

This 3-day workshop will offer an overview of the most popular uses of MSEM. MSEM is the most general and flexible analytic framework within the broader general linear model. Single level SEM and basic MLM can be understood as special cases of MSEM. However, within MSEM investigators can flexibly move between traditional SEM and MLM and combine the strengths of each (e.g., latent variables, nested data, random effects). In addition, MSEM software offers several advantageous features, like latent decomposition of Level 1 variables into Level 1 and Level 2 variance portions, standardized effects for multilevel models, the ability to accommodate observation level missing data, among others.

Most behavioral and health sciences use the general linear model (e.g., regression, ANOVA) and some extensions (e.g., SEM, MLM). But despite the flexibility of these techniques, they are often taught as siloed and not linked to each other, and so researchers often hit roadblocks in their thinking about their conceptual problems. MSEM is the most flexible and general technique and will remove artificial barriers to thinking through conceptual questions with data.

What you’ll learn

  • Perspective – Remove the blinders imposed by specific statistical techniques and start seeing the full landscape of modeling possibilities

  • Psychometrics – Learn the basics of psychometrics for multilevel data (e.g., ecological momentary assessment questionnaires)

  • Tricks of the trade – Learn hard won insights of doing MSEM in real data from someone who has applied these techniques in dozens of data sets

  • Depth of understanding – Even experienced SEM and MLM users will understand these basic techniques better through using them together in MSEM

Syllabus

Day 1

  • Multilevel Confirmatory Factor Analysis

  • Evaluating Model Fit

  • Multilevel Psychometrics

Day 2

  • Introduction to multilevel path analysis

  • Using the Bayes Estimator

  • Multilevel Mediation

  • Slopes as predictors models

Day 3

  • Full MSEM

  • Dynamic SEM (DSEM)

Registration Options

Multilevel Structural Equation Modeling

  • Professional
  • $824
  • Baseline Price for Faculty,
    Staff, and Other Professionals
  • Click Register Below
  • Trainee
  • $824 $549
  • 33% Discount for
    Students and Postdocs
  • Use code "TRAINEE" at Checkout

Combo 1: Multilevel Structural Equation Modeling + Transitioning to Mplus of Multilevel Modeling

  • Professional
  • $1423 $1139
  • Baseline Price for Faculty,
    Staff, and Other Professionals
  • 20% Combination Discount
  • Click Register Below
  • Trainee
  • $1139 $759
  • 33% Discount for
    Students and Postdocs
  • 20% Combination Discount
  • Use code "TRAINEE" at Checkout

Combo 2: Multilevel Structural Equation Modeling + Designing Ambulatory Assessment Studies

  • Professional
  • $1423 $1139
  • Baseline Price for Faculty,
    Staff, and Other Professionals
  • 20% Combination Discount
  • Click Register Below
  • Trainee
  • $1139 $759
  • 33% Discount for
    Students and Postdocs
  • 20% Combination Discount
  • Use code "TRAINEE" at Checkout

Note: All registration options include access to video recordings of the workshop for continued access and for self-paced learning for those who cannot attend live. Recordings will be uploaded to the workshop page shortly after each day ends.

 FAQs

  • Investigators interested in moving beyond typical boundaries imposed by simpler modeling frameworks, especially those interested in more sophisticated interrogation of intensive longitudinal data.

  • Advanced

  • Prior knowledge of or experience with multilevel modeling or with structural equation models is necessary and expected. Ideally learners would be conversant with both multilevel and structural equation modeling. Basic knowledge of the Mplus modeling software is encouraged, but not absolutely necessary if fundamentals of SEM and MLM are strong.

  • Mplus with the multilevel add on is required.

  • Participants will be provided with annotated example code, practice data, and example write-up.