探索性结构方程建模(ESEM)是在测量模型部分使用了类似于EFA模型的SEM。作为一种高级统计方法,ESEM整合了EFA和CFA两种因子分析方法的功能和优点。通过ESEM,研究者既可以灵活地探索因子结构,又可以系统地验证因子模型,为潜变量的关系分析提供更适宜的测量模型。ESEM已经在某些社科领域的研究中得到应用,是一种值得推介的因子分析方法。ESEM的具体应用问题,例如因子旋转方法的选用、测验信度评价等,仍有待探讨。
Exploratory structural equation modeling (ESEM) is an approach in which an exploratory factor analysis measurement model with rotations is a part of an SEM. As a kind of advanced statistical methods, ESEM integrates the functions and advantages of exploratory and confirmatory factor analysis, two kinds of factor analysis methods. Using ESEM, researchers can flexibly explore the factor structure, systematically test the factor model, and provide more suitable measurement model for analyzing the relationship of latent variables. ESEM has been applied in the researches of some disciplines in social science. It is a method deserved of recommendation for factor analysis. Some issues on ESEM, such as the choice of rotation criterion and the methods for estimating test reliability, require to be addressed in the future.