Longitudinal Data Analysis: Topics

The table below provides an outline of the topics covered in the LDA workshop. LDA encompasses random effect models for change, latent growth curve, latent change score models, and survival analysis and other models for event occurrence. Briefly, day 1 introduces the kinds of research questions approached using LDA; an introduction to Mplus software; using Mplus software; and the importance of workflow; finally, LDA models are introduced. By the end of day 1 we have demonstrated that the latent growth curve model can be specified in a way that produces the same parameter estimates as mixed effect models for change (e.g., SAS/PROC MIXED, Stata/xtmixed). Day 2 expands on latent growth curve models, discusses data handling issues, model fit assessment, and latent classes for growth. Day 3 covers applications and more advanced topics.

Day 1

Introduction and Overview Longitudinal Data Analysis

  • Workshop Objectives

  • Content Covered in Workshop

  • Resources to Continue Learning After Workshop

Introduction to Longitudinal Data Analysis (LDA)

  • Alternatives and Challenges

  • Problems with Change Scores

  • Modeling as a Research Paradigm

  • Latent Growth Curve Modeling (LGM)

  • Thinking about and modeling time

Orientation to General Latent Variable Modeling Framework

  • Path Diagram Notation

  • Covariance Structure Modeling Perspectives on Latent Variables

  • Orientation to Mplus Modeling

    • How to write an Mplus command file

    • Example Data Set Overview

    • Getting Data into Mplus (Using SAS, SPSS, Stata, R; PC and MAC)

    • Workflow and Reproducibility

  • Confirmatory Factor Analysis (CFA)

  • Growth Curve Modeling

    • Compare with SAS/PROC Mixed

Day 2

Latent Growth Curve Models

  • LGM Model Specification

  • LGM with covariates

    • Time Invariant Covariates

    • Time Varying Covariate

  • Using Mplus for LGM

    • Mplus Command Syntax

    • Missing Data Handling

    • Model Fit Assessment

    • Data Handling: Centering Covariates

    • Non-Linear Development

Growth Mixture Modeling (GMM)

  • Latent Class Analysis (LCA) for Growth

    • Assessing Fit

    • Determining the number of classes

  • Exploratory and Confirmatory LCA

  • GMM with Covariates

  • Importance of Theory and Hypotheses for Driving Data Analysis

Multilevel Model Approach to Growth Curve Modeling with Mplus

Multiple Indicator Growth Curve Model

Modeling Retest Effects and Other Methods Artifacts

Day 3

Applications

  • Knownclass Mixture Model

  • LGM in Randomized Controlled Trials

    • Acceleration/Deceleration

    • Punctuation

Advanced and alternative models

  • Latent Change Score Models

    • Dual Change Score Model

  • Strategies for Building Complex Models

    • Accelerated Longitudinal Designs

    • Multilevel LGM

    • Parallel Process LGM

    • Multiple Indicator LGM with Measurement Non-Invariance

    • Joint Survival and Growth Models

Resources to Continue Learning After Workshop