Can Target Oblique Rotation Recover Factor Correlations of a Bifactor Structure?

• In Event: Factor Analysis: Model Specification and Fit

Abstract

Bifactor model assumes the correlations between the primary and subdomain factors and among the subdomain factors themselves are orthogonal. It is a flawed statement because items can never be perfectly written to measure the specified constructs so that no correlations exist among all the constructs. Orthogonal assumption of the bifactor model can result in a loss of information and can lead to distorted item parameter estimates in the bifactor solution. Target oblique rotation can relax the constraint of the bifactor model. Little has been investigated in the accuracy of the factor correlation recovery using the target rotation method. The purpose of this study was to examine if factor correlations of a bifactor model can be recovered using target oblique rotation.

Authors

• Chunmei Zheng, Pearson

• Paul Johnson, The University of Kansas