Iteration of Partially Specified Target Matrices: Applications in Exploratory and Bayesian Confirmatory Factor...

Area of Science:

• Psychometrics
• Statistical Modeling
• Data Analysis

Background:

• Traditional target rotation requires pre-specified target matrices.
• Iterated target rotation (ITR) offers an alternative by iteratively updating the target matrix.

Purpose of the Study:

• To introduce and evaluate the Iterated Target Rotation (ITR) algorithm.
• To compare ITR's performance against standard analytic rotations.
• To explore ITR's application in Bayesian confirmatory factor analysis (BCFA).

Main Methods:

• Monte Carlo simulations were used to assess ITR performance with varying factor structure complexity.
• ITR was applied to a rater-report alexithymia measure within a BCFA framework.
• The study evaluated ITR's accuracy and its utility in specifying empirically informed priors.

Main Results:

• ITR is particularly effective for analyzing complex factor structures with multiple cross-loadings.
• The initial rotation method used to establish a target matrix did not significantly impact ITR's accuracy.
• ITR can be utilized to derive empirical priors for BCFA, though challenges in prior specification and model fit assessment exist.

Conclusions:

• Iterated target rotation (ITR) provides a valuable tool for factor analysis, especially with complex data.
• ITR facilitates the development of empirically grounded priors in Bayesian confirmatory factor analysis.
• Further research is needed to refine prior specification and model fit evaluation within ITR-informed BCFA.