贝叶斯多组比较——渐近测量不变性

在行为科学研究领域中,检验测量工具的测量不变性是进行群体差异比较的前提。目前,多组验证性因子分析(多组CFA)方法被广泛用于检验测量不变性,但是它对跨组等值的限制过于严格,在实际应用中常常存在大量局限。贝叶斯渐近测量不变性方法基于贝叶斯思想的优良特性,放宽了传统多组CFA方法对跨组差异的严格限制,避免了传统方法的问题,具有较高的应用价值。文章详细介绍了贝叶斯渐近测量不变性方法的原理及优势,同时通过实例展示了渐近测量不变性方法在Mplus软件中的具体分析过程。

Bayesian Multiple-group Analysis:Approximate Measurement Invariance

In the behavioral science,the comparison of multiple groups under the latent variable framework is popular.Measurement invariance(MI)is a pre-requisite for such multiple-group comparison.Multiple-group Confirmatory Factor Analysis(Multiple-group CFA)is the most commonly used approach for testing measurement invariance.In traditional multi-group CFA,strict invariant constraints are imposed on measurement parameters across groups.However,due to the complexity of modeling multi-group data,these strict constraints are unrealistic in real data analysis and can easily lead to poor model fitting.In fact,scalar invariance is almost unachievable in practice.The Bayesian approximate MI proposed by Muthén and Asparouhov(2013)compensates for these limitations to some extent by providing a zero-mean,small-variance prior for the differences in measurement parameters.It allows for small differences between groups and avoids the problems caused by strict restrictions in classical method,such as poor model fitting,awkward model modifications and higher Type I error rate.These strengths make this new approach a more suitable method in the practical research.This paper introduced the principles and advantages of the approximate MI approach by comparing it with the classical multi-group CFA method.Besides,a real data set was analyzed to demonstrate the validity and application of this approach using Mplus.