Assessing the Robustness of Growth Mixture Models with Distal Outcomes Across Different Time Coding Approaches

Growth mixture models (GMMs) with distal outcomes identify groups following different trajectories and examine group differences in subsequent outcomes. Although time coding is often treated as a matter of parameter interpretation, it may also affect model estimation, class recovery, and distal-outcome inference. We examined this sensitivity through two empirical illustrations and simulation studies varying time coding, data conditions, and population trajectory patterns. Alternative codings sometimes selected different numbers of classes. Even with the number of classes fixed, fitted models differed in convergence and attained log-likelihood solutions, accompanied by differences in class recovery and distal-outcome estimates. These discrepancies were most pronounced when the population contained more classes. Moreover, the relative performance of each coding varied across trajectory patterns, with no coding performing consistently best. Researchers should therefore select a primary coding based on substantive interpretation and examine sensitivity across alternative codings, particularly when many classes are considered, or estimation is unstable.

Authors

Publication Details

Journal
Figshare
Published
2026-11-19
DOI
https://doi.org/10.6084/m9.figshare.35068189
Primary Topic
Psychometric Methodologies and Testing
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Assessing the Robustness of Growth Mixture Models with Distal Outcomes Across Different Time Coding Approaches

Zsuzsa Bakk, Mark de Rooij, Ethan Michael McCormick, Yuqi Liu
Figshare
Psychometric Methodologies and Testing
article

Assessing the Robustness of Growth Mixture Models with Distal Outcomes Across Different Time Coding Approaches

Zsuzsa Bakk, Mark de Rooij, Ethan Michael McCormick, Yuqi Liu
article en

Abstract

Growth mixture models (GMMs) with distal outcomes identify groups following different trajectories and examine group differences in subsequent outcomes. Although time coding is often treated as a matter of parameter interpretation, it may also affect model estimation, class recovery, and distal-outcome inference. We examined this sensitivity through two empirical illustrations and simulation studies varying time coding, data conditions, and population trajectory patterns. Alternative codings sometimes selected different numbers of classes. Even with the number of classes fixed, fitted models differed in convergence and attained log-likelihood solutions, accompanied by differences in class recovery and distal-outcome estimates. These discrepancies were most pronounced when the population contained more classes. Moreover, the relative performance of each coding varied across trajectory patterns, with no coding performing consistently best. Researchers should therefore select a primary coding based on substantive interpretation and examine sensitivity across alternative codings, particularly when many classes are considered, or estimation is unstable.

Figshare
Psychometric Methodologies and Testing
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon