Comparative framework for longitudinal models differing in change processes, temporal structure, and measurement design

Research Method · All research methods
Evidence · 5 claims (4 for, 1 mixed) · 3 studies (3 associational), q1–q2 · 0 of 3 report an effect size · 5 claims rest on one study

Description

Beyond responding to the commentary, the authors "present a generalizable framework for comparing longitudinal models that differ in their assumptions about change processes, temporal structure, and measurement design." The framework organizes model comparison around three axes: what change process a model assumes, how it represents temporal structure (e.g., lag0 and lag1 effects), and what measurement design it presumes. It is used in the article to reanalyze the simulated and PALMA data and evaluate the extended REM against alternatives.

Accounts

• Comparative framework for longitudinal models differing in change processes, temporal structure, and measurement design: Beyond responding to the commentary, the authors "present a generalizable framework for comparing longitudinal models that differ in their assumptions about change processes, temporal structure, and measurement design." The framework organizes model comparison around three axes: what change process a model assumes, how it represents temporal structure (e.g., lag0 and lag1 effects), and what measurement design it presumes. It is used in the article to reanalyze the simulated and PALMA data and evaluate the extended REM against alternatives. (Fernando Núñez-Regueiro et al. (2025))
• Juxtapose results from competing panel models rather than selecting a single best model by goodness-of-fit: The article recommends that longitudinal model evaluation rest on substantive interpretation alongside fit indices, following Muthén and Asparouhov's advice. Because alternative models may be formally equivalent or hard to distinguish by fit, "conclusions should be based on the juxtaposition of different results rather than selecting a single "best" model based on goodness-of-fit" when competing models each support a priori theory-based hypotheses. Researchers should report results from competing models and consider more parsimonious constrained versions when convergence problems arise. (Marsh et al. (2024))
• Test alternative longitudinal models and juxtapose their results and interpretations before drawing causal conclusions: The article endorses Marsh et al.'s (2024) recommendation that researchers "test alternative models and juxtapose their results and interpretations". When findings from alternative models converge, causal conclusions are corroborated; when they diverge, "caution is advised and causal conclusions appear premature". The authors demonstrate this by fitting RI-CLPM, LCSM, ESDM and STARTS models to the same simulated data and finding contradictory positive, negative, and null effects. (Kimmo Sorjonen et al. (2025))

Claims

• The extended REM was robust to several alternative modeling specifications proposed by Sorjonen et al. [+M]
• Random-intercept and lag2 panel models rest on different assumptions and offer alternative perspectives on controlling covariates [~W]
• RI-CLPM on simulated PALMA-like data shows positive reciprocal cross-lagged effects between within-individual math achievement and self-concept [+M]
• LCSM on the same simulated PALMA-like data shows negative cross-lagged effects, contradicting the reciprocal effects model [+M]
• Validation with simulated TRAIN-like data replicates the divergence: RI-CLPM positive, LCSM negative, ESDM no-direct-effects [+M]

Related Research Methods

• State–trait temporality framework for child engagement measurement

Key Sources

• Fernando Núñez-Regueiro, Herbert W Marsh, Reinhard Pekrun, Oliver Lüdtke, Jiesi Guo. (2025). The Reciprocal Effects Model is Robust to Alternative Modeling Specifications: A Response to Sorjonen et al., 2025. Educational Psychology Review, 37 (3), pp.79. https://doi.org/10.1007/s10648-025-10059-7
• Marsh, H. W., Guo, J., Pekrun, R., Lüdtke, O., & Núñez-Regueiro, F. (2024). Cracking Chicken-Egg Conundrums: Juxtaposing Contemporaneous and Lagged Reciprocal Effects Models of Academic Self-Concept and Achievement's Directional Ordering. Educational Psychology Review, 36. https://doi.org/10.1007/s10648-024-09887-w
• Kimmo Sorjonen, Bo Melin, Gustav Nilsonne. (2025). Inconclusive Evidence for a Prospective Effect of Academic Self-Concept on Achievement: A Simulated Reanalysis and Comment on Marsh et al. (2024). Educational Psychology Review. https://doi.org/10.1007/s10648-025-10008-4

Page metadata

Field	Value
type	research-method
id	comparative-framework-for-longitudinal-models-differing-in-change-processes-temporal-struc
title	Comparative framework for longitudinal models differing in change processes, temporal structure, and measurement design
description	Beyond responding to the commentary, the authors "present a generalizable framework for comparing longitudinal models that differ in their assumptions about change processes, temporal structure, and measurement design." The framework organizes model comparison around three axes: what change process
status	review
generated	by process:settle-candidates on 2026-10-19
trust	unverified — no one has confirmed this page's content

Frontmatter is the page's machine-readable record — see the Schema & Guide.