Alleviating estimation problems in small sample structural equation modeling—A comparison of constrained maximum likelihood, Bayesian estimation, and fixed reliability approaches.

Abstract

Small sample structural equation modeling (SEM) may exhibit serious estimation problems, such as failure to converge, inadmissible solutions, and unstable parameter estimates. A vast literature has compared the performance of different solutions for small sample SEM in contrast to unconstrained maximum likelihood (ML) estimation. Less is known, however, on the gains and pitfalls of different solutions in contrast to each other. Focusing on three current solutions-constrained ML, Bayesian methods using Markov chain Monte Carlo techniques, and fixed reliability single indicator (SI) approaches-we bridge this gap. When doing so, we evaluate the potential and boundaries of different parameterizations, constraints, and weakly informative prior distributions for improving the quality of the estimation procedure and stabilizing parameter estimates. The performance of all approaches is compared in a simulation study. Under conditions with low reliabilities, Bayesian methods without additional prior information by far outperform constrained ML in terms of accuracy of parameter estimates as well as the worst-performing fixed reliability SI approach and do not perform worse than the bestperforming fixed reliability SI approach. Under conditions with high reliabilities, constrained ML shows good performance. Both constrained ML and Bayesian methods exhibit conservative to acceptable Type I error rates. Fixed reliability SI approaches are prone to undercoverage and severe inflation of Type I error rates. Stabilizing effects on Bayesian parameter estimates can be achieved even with mildly incorrect prior information. In an empirical example, we illustrate the practical importance of carefully choosing the method of analysis for small sample SEM.

Comparing the Robustness of the Structural after Measurement (SAM) Approach to Structural Equation Modeling (SEM) against Local Model Misspecifications with Alternative Estimation Approaches

“…The inequality constraint par > 0.01 was used where par was a loading or variance parameter. Previous research has shown that constrained (ML) estimation is preferable to unconstrained (ML) estimation in small samples because it solves convergence issues and substantially reduces variability in estimates [62,63].…”

Measurement matters: A commentary on the state of the science on patient reported outcome measures (PROMs) in autism research

“…Notably, many of the more advanced statistical approaches indeed require large samples, which may not be feasible given constraints on resources or the specific autistic sub‐population under study (e.g., childhood disintegrative disorder, autism associated with rare neurogenetic syndromes). In such cases, rather than attempt to utilize “large‐sample” statistical techniques, researchers are advised to use methods appropriate for the sample size, which may include classical test theory approaches (DeVellis, 2006; Nolte et al, 2019), Rasch models (Cleanthous et al, 2019; Petrillo et al, 2015), or Bayesian structural equation modeling (Smid et al, 2020; Ulitzsch et al, 2023) to analyze available data.…”

Mixture of longitudinal factor analyzers and their application to the assessment of chronic pain

“…However, we wanted to inform the readers about this issue, which is present in the structural equation model framework as a whole. When only very few observations are present in a mixture component, the component parameters should be interpreted with caution 20 . Models with less components or including more observations should also be considered.…”