Modeling Misspecification as a Parameter in Bayesian Structural Equation Models

Educational and Psychological Measurement, Ahead of Print.
Accounting for model misspecification in Bayesian structural equation models is an active area of research. We present a uniquely Bayesian approach to misspecification that models the degree of misspecification as a parameter—a parameter akin to the correlation root mean squared residual. The misspecification parameter can be interpreted on its own terms as a measure of absolute model fit and allows for comparing different models fit to the same data. By estimating the degree of misspecification simultaneously with structural parameters, the uncertainty about structural parameters reflects the degree of model misspecification. This results in a model that produces more reliable inference than extant Bayesian structural equation modeling. In addition, the approach estimates the residual covariance matrix that can be the basis for diagnosing misspecifications and updating a hypothesized model. These features are confirmed using simulation studies. Demonstrations with a variety of real-world examples show additional properties of the approach.

Adjusting to the “New Normal”: How were mental health and self-care affected in patients with diabetes mellitus 1 year into the COVID-19 crisis? A longitudinal study

Journal of Health Psychology, Ahead of Print.
This study aimed to assess the long-term effect of the pandemic on mental health and self-care parameters in patients with diabetes during the COVID-19 pandemic in Brazil. After 18 months of pandemic, 118 participants remained in the study (mean age of 56.6 ± 13.4 years, 66.7% were women). We observed no change in the scores for mental health disorders screening. Regarding self-care, patients with type 1 diabetes showed an improvement in the adherence score compared to those found at the beginning of the pandemic (variation + 3.5 (−6.0 to +15.8) points, p = 0.02), and also compared to those with type 2 diabetes. Although the pandemic have negatively affected many people’s mental health, especially in those with chronic diseases, our results show that patients with diabetes may have developed good coping and adaptive strategies to maintain diabetes control and symptom pattern of mental health disorders over the course of the pandemic.

A Note on Comparing the Bifactor and Second-Order Factor Models: Is the Bayesian Information Criterion a Routinely Dependable Index for Model Selection?

Educational and Psychological Measurement, Ahead of Print.
This note demonstrates that the widely used Bayesian Information Criterion (BIC) need not be generally viewed as a routinely dependable index for model selection when the bifactor and second-order factor models are examined as rival means for data description and explanation. To this end, we use an empirically relevant setting with multidimensional measuring instrument components, where the bifactor model is found consistently inferior to the second-order model in terms of the BIC even though the data on a large number of replications at different sample sizes were generated following the bifactor model. We therefore caution researchers that routine reliance on the BIC for the purpose of discriminating between these two widely used models may not always lead to correct decisions with respect to model choice.