The Detection and Modeling of Direct Effects in Latent Class Analysis.

Leiden Repository

The Detection and Modeling of Direct Effects in Latent Class Analysis.

Type: Article / Letter to editor
Title: The Detection and Modeling of Direct Effects in Latent Class Analysis.
Author: Janssen, J.H.M.Laar, S. vanRooij, M.J. deKuha, J.Bakk, Z.
Journal Title: Structural Equation Modeling: A Multidisciplinary Journal
Issue: 2
Volume: 26
Start Page: 280
End Page: 290
Pages: 11
Issue Date: 2019
Keywords: Direct effects, Latent class analysis, Mixture models, Residual statistics
Abstract: Several approaches have been proposed for latent class modeling with external variables, including one-step, two-step, and three-step estimators. However, very little is known yet about the performance of these approaches when direct effects of the external variable to the indicators of latent class membership are present. In the current article, we compare those approaches and investigate the consequences of not modeling these direct effects when present, as well as the power of residual and fit statistics to identify such effects. The results of the simulations show that not modeling direct effect can lead to severe parameter bias, especially with a weak measurement model. Both residual and fit statistics can be used to identify such effects, as long as the number and strength of these effects is low and the measurement model is sufficiently strong.
Uri: https://www.tandfonline.com/doi/full/10.1080/10705511.2018.1541745
Handle: http://hdl.handle.net/1887/70527
 

Files in this item

Description Size View
application/pdf Text_IR_LEI 1.005Mb View/Open Full text at publisher site

This item appears in the following Collection(s)