In this study, we propose an approach to further improve the performance of ULMC by adding a single identifying variable together with the usage of Rindskopf’s (1983) reparameterization of the CFA model for error variances and implicit constraints for correlations, based on the multi-dimensional ULMC model. The chi-square test for model fit and the chi-square difference test are used to test for the existence of CMV and to determine the number of method dimensionality. The simulation results have indicated that, given adequate sample size, the ULMC technique with the proposed approach performs well in detecting CMV and in yielding acceptable corrected trait loadings and trait correlations. Moreover, its performance in CMV correction is superior to the performance of the CFA marker technique. Therefore, we recommend that the ULMC technique with the proposed approach be used for CMV detection and correction."
|主出版物標題||Accepted for Presentation at the 2020 Academy of Management Annual Meeting – Research Methods Division, Vancouver, BC, Canada, August 7-11, 2015. (did not attend because of the coronavirus (COVID-19) pandemic.)|
|出版狀態||Published - 2020|