Paper Summary
Share...

Direct link:

Finding Optimal Factor Analytic Representations of the Structure of Personality (Poster 5)

Thu, April 13, 4:40 to 6:10pm CDT (4:40 to 6:10pm CDT), Hyatt Regency Chicago, Floor: East Tower - Exhibit Level, Riverside West Exhibition Hall

Abstract

The purpose of this study is to examine differences in average model fit indices between selected pairs of models to compare relative improvements in fit gained by altering estimation procedures and allowing non-zero off-target loadings for correlated factor, hierarchical, and bifactor models of personality using a sample of 447,500 respondents who completed IPIP-NEO-120 questionnaire. Results across the three types of investigated factor models highlighted the importance of controlling for scale coarseness, allowing small off-target loadings when using maximum likelihood (ML) and robust weighted least squares estimation (WLSMV), and including informative priors when using Bayesian estimation. Although the best fits were yielded by bifactor, excellent fits were found for each type of model when properly configured.

Authors