Victoria Savalei
@vsavalei.bsky.social
Quantitative Psychologist at the University of British Columbia
Reposted by Victoria Savalei
New research by Wendy Higgins, @vsavalei.bsky.social, Vince Polito, and me shows that the Reading the Mind in the Eyes Test scores have poor structural properties. journals.sagepub.com/doi/10.1177/...
Reading the Mind in the Eyes Test Scores Demonstrate Poor Structural Properties in Nine Large Non-Clinical Samples - Wendy C. Higgins, Victoria Savalei, Vince Polito, Robert M. Ross, 2025
The Reading the Mind in the Eyes Test (RMET) is widely used in clinical and non-clinical research. However, the structural properties of RMET scores have yet to...
journals.sagepub.com
April 1, 2025 at 5:15 AM
New research by Wendy Higgins, @vsavalei.bsky.social, Vince Polito, and me shows that the Reading the Mind in the Eyes Test scores have poor structural properties. journals.sagepub.com/doi/10.1177/...
This was a weird paper to write! The sheer number of EFA structures that was deemed 'passing' by CFA fit indices made me realize that perfect simple structure is really more of an assumption than a real tested proposition in CFA.
February 20, 2025 at 5:23 PM
This was a weird paper to write! The sheer number of EFA structures that was deemed 'passing' by CFA fit indices made me realize that perfect simple structure is really more of an assumption than a real tested proposition in CFA.
Thus, good fit of a CFA model only rules out that the true model is an EFA model with severely violated constraints on the ratios of main and cross-loadings, but it does not rule out most other plausible EFA structures.
February 20, 2025 at 5:23 PM
Thus, good fit of a CFA model only rules out that the true model is an EFA model with severely violated constraints on the ratios of main and cross-loadings, but it does not rule out most other plausible EFA structures.
In fact, fit indices are non-monotone functions of the number of positive cross-loadings in the true model, so that more cross-loadings in the true model does not necessarily translate into worse fit of the CFA model! The relationship becomes monotone only when cross-loadings are of mixed sign.
February 20, 2025 at 5:23 PM
In fact, fit indices are non-monotone functions of the number of positive cross-loadings in the true model, so that more cross-loadings in the true model does not necessarily translate into worse fit of the CFA model! The relationship becomes monotone only when cross-loadings are of mixed sign.
There could be a word missing or it could be a hidden message!
December 18, 2024 at 1:56 AM
There could be a word missing or it could be a hidden message!
excited to read these!
October 25, 2024 at 4:54 PM
excited to read these!