Longitudinal Data Analysis

14.03.2023:

Prof. Dr. Craig Enders: Longitudinal Modeling and Missing Data Handling In Blimp

Missing data are a ubiquitous feature of nearly all longitudinal modeling applications, arising through participant non-response, attrition, and sometimes even by design. Failure to account appropriately for missing values when conducting statistical analyses can result in badly biased estimates and incorrect inferences about the relationships under study. Longitudinal Modeling and Missing Data Handling with Blimp is a full-day workshop focused on Bayesian estimation and multiple imputation, as implemented in the Blimp software application. These procedures are advantageous because they use all available data and make realistic assumptions about the cause of missingness; estimates and significance tests are therefore valid in a broader range of situations than historical methods such as deleting incomplete data records. The purpose of this workshop is to provide participants with foundational knowledge about the application of Bayesian estimation and multiple imputation to longitudinal data analyses. To this end, the workshop will include a mix of theoretical information, practical tips, and computer demonstrations involving real world data sets. A review of mixed (multilevel) models for longitudinal data will be provided, but familiarity with this topic will be beneficial. Workshop topics are listed below.

Review of mixed models for longitudinal data analyses

Review of Rubin’s missing data mechanisms

Overview of factored regression specifications

Overview of Bayesian estimation and multiple imputation

Missing data handling for linear and curvilinear trends

Dealing with incomplete covariates in longitudinal models

Modeling missing not at random processes

Missing data handling for three-level models

 

Dr. Christian Geiser: Latent State-Trait Modeling with Mplus

In this applied workshop, Christian Geiser provides an introduction to latent state-trait modeling in the Mplus software. The workshop covers basic and advanced models and methods of longitudinal confirmatory factor analysis. We will discuss longitudinal measurement invariance testing and analyze models for separating trait, state residual, method, and measurement error components. Participants can bring their own laptop with the demo version of Mplus to follow the data examples