ISSN 1671-3710
CN 11-4766/R
主办:中国科学院心理研究所
出版:科学出版社

心理科学进展 ›› 2004, Vol. 12 ›› Issue (2): 231-239.

• • 上一篇    下一篇

测量等价性的概念及其判定条件

白新文,陈毅文   

  1. 中国科学院心理研究所,北京 100101
  • 收稿日期:2003-08-18 修回日期:1900-01-01 出版日期:2004-03-15 发布日期:2004-03-15
  • 通讯作者: 陈毅文

MEASUREMENT EQUIVALENCE: CONCEPTIOIN AND TEST CONDITIONS

Bai Xinwen,Chen Yiwen   

  1. Institute of Psychology, Chinese Academy of Sciences, Beijing 100101
  • Received:2003-08-18 Revised:1900-01-01 Online:2004-03-15 Published:2004-03-15
  • Contact: Chen Yiwen

摘要: 测量等价性指的是,应用量表进行测量时,当观测变量和潜在特质之间的关系在相比较的各个组之间等同时,就称该量表具备测量等价性。特别地,来自不同群体但在潜在特质上得分相等的个体,他们观测变量的得分也应该相等。测量工具满足测量等价性的要求是进行组间差异比较的前提条件。该文首先明确了测量等价性的概念及其研究历史,然后阐述了测量等价性的重要性以及对测量等价性分析的必要性,进而讨论了在结构方程模型中测量等价性所要满足的5个条件,最后列举了模型优劣判定的拟合度指数

关键词: 测量等价性, 结构方程模型, 拟合度

Abstract: Equivalent measurement is obtained when the relations between observed test scores and the latent attribute measured by the test are identical across subpopulations. In particular, individuals with equal standings on the latent trait, but sampled from different subpopulations, should have the same expected observed score. Measurement equivalence serves as the precondition to conducting substantive cross-group difference comparisons. This article introduces the conception of measurement equivalence, the brief history of the research, and the necessity of measurement equivalence; then lists the conditions of measurement equivalence within the framework of structural equation modeling (SEM); also presents the goodness-of-fit indices which are used to justify the goodness of fit of the models.

Key words: measurement equivalence, structural equation modeling (SEM), goodness-of-fit

中图分类号: