A Monte-Carlo Study of Measurement Equivalence/Invariance between Factor Loadings

• Main Authors: Chen, Kuan-Chih, 陳冠志
• Other Authors: Yang, Chih-Chien
• Format: Others
• Language: zh-TW
• Published: 2006
• Online Access: http://ndltd.ncl.edu.tw/handle/24412555664452527006

碩士 === 國立臺中教育大學 === 教育測驗統計研究所 === 94 === 　　In the process of testing measurement equivalence of confirmatory factor analysis between multiple groups, one of the most critical steps is to test the equivalences between factor loadings. This study is focus on this step to discuss the steadiness of test the equivalences between factor loadings. This study is use the Monte Carlo method that is design the 10 types of sample size and 15 types of factor loading difference. And this study is design two methods to test the equivalences between factor loadings. One is the “one by one”, and the other is “all in one”. 　　The results of this study are as follows: In this study, use the “one by one” method to test the equivalences between factor loadings is better then the “all in one” method when the factor loading difference less then 0.1. When sample size less then 500, this study is suggest to use the “one by one” method to test the equivalences between factor loadings. When sample size more then 500, this study is suggest to use the “all in one” method to test the equivalences between factor loadings before use the “one by one” method to test the equivalences between factor loadings.