Improper Solutions in Structural Equation Models

By Feinian Chen, Kenneth A. Bollen, Pamela Paxton, P. Curran, James B. Kirby. 2001

549 citations recorded by the source corpus.

Subfields: Management Science and Operations Research

View Paper: https://www.semanticscholar.org/paper/da16b23832a830b51392870fb9af535ad36c00e6

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Management Science and Operations Research

Citations in the corpus

Unintended Consequences of using Tests to Improve Learning: How Improvement-Oriented Resources Heighten Conceptions of Assessment as School Accountability (2009, background): “…value of the negative error variance so that there is a strong likelihood that the true value of the error variance includes zero and when the model has been shown to work properly elsewhere it is valid to set the negative error variance to a small positive value (e.g., .005) (Chen et al., 2001).”
Distributional assumptions of growth mixture models: implications for overextraction of latent trajectory classes. (2003, background): “Unless convergence was of explicit interest, nonconverged and improper solutions were excluded from the analyses because such solutions are rarely interpreted in practice ( Chen, Bollen, Paxton, Curran, & Kirby, 2001 ).”
Factor Analysis with Ordinal Indicators: A Monte Carlo Study Comparing DWLS and ULS Estimation (2009, background): “Notice that improper or nonconvergent solutions are of no use to the applied researcher ( Chen, Bollen, Paxton, Curran, & Kirby, 2001 ).”
An Empirical Comparison of the Efficacy of Covariance-Based and Variance-Based SEM (2009, methodology): “In line with a previous research, we define improper solutions as those estimates that would be impossible (or implausible) for the corresponding parameters (e.g., Bollen, 1987; Chen et al., 2001) and consider an ML-based CBSEM solution as proper if the iteration process converges to some solution a”
CORRELACIONES POLICÓRICAS Y TETRACÓRICAS EN ESTUDIOS FACTORIALES EXPLORATORIOS Y CONFIRMATORIOS POLYCHORIC AND TETRACHORIC CORRELATIONS IN EXPLORATORY AND CONFIRMATORY FACTORIAL STUDIES (2015, unknown): “Algunos teóricos en sus estudios concluyen que la presencia de ecuaciones infractoras no sería un factor determinante para el ajuste de un modelo y que su no tratamiento es una alternativa a tener en cuenta (Chen, Bollen, Paxton, Curran y Kirby, 2001; Gerbing y Anderson, 1987).”
Testing Negative Error Variances (2012, methodology): “The likelihood ratio test rejects too frequently, but seems to perform the best (Chen et al. 2001:498-99).”