cresstucla.bsky.social
Center for Research on Evaluation, Standards and Student Testing at @uclaseis.bsky.social
https://cresst.org/
16 posts
74 followers
54 following
Regular Contributor
comment in response to
post
Bonifay, W., Cai, L., Falk, C. F., & Preacher, K. J. (2025). Reassessing the fitting propensity of factor models. Psychological Methods. doi.org/10.1037/met0...
comment in response to
post
This work underscores the need for more nuanced approaches to model selection in psychological research.
comment in response to
post
3. FP varies significantly across factor models, highlighting the importance of considering parametric and configural complexity in model evaluation.
comment in response to
post
Key findings:
1. Bifactor models with fewer specific factors have higher FP, making them more likely to overfit data.
2. Exploratory factor analysis (EFA) models outperform confirmatory bifactor models in fitting diverse data, even with the same parameters.
comment in response to
post
They challenge the idea that the number of parameters solely determines model complexity.
Their study reveals that a model’s functional form (how parameters are arranged) is critical to fit diverse data patterns, impacting its generalizability and interpretability.
comment in response to
post
Jia sends good wishes from Beijing!