报告题目：Corrected Test Statistics for Structural Equation Modeling with Large p and Small N
报告人单位：University of Notre Dame
报告内容：Survey data typically contain many variables. Structural equation modeling (SEM) is commonly used in analyzing such data. The most widely used statistic for evaluating the adequacy of a SEM model is the likelihood ratio statistic Tml. When the number of observations (N) is large and the number of items or variables (p) is small, Tml performs reasonably well for normally distributed data or with certain violation to normality. However, in practice, p can be rather large while N is always limited due to not having enough participants. Empirical results show that Tml rejects the correct model close to 100% when p is relatively large. Better statistics are needed for SEM with large p and/or small N. Following the principle of Bartlett correction, this paper proposes an empirical approach to correct Tml so that the mean of the resulting statistic approximately equals the degrees of freedom of the nominal chi-square distribution. Results show that empirically corrected statistics follow the nominal chi-square distribution much more closely than previously proposed corrections to Tml, and they control type I errors reasonably well whenever N is greater than 50 or 2p. The corrected statistics are also much better candidates for defining fit indices than Tml.
报告人简介：袁克海博士，美国圣母大学心理系教授。他在UCLA获得博士学位，研究兴趣包括结构方程模型、中介分析和调节分析、多水平模型、混合模型、项目反映理论、元分析等。他在Journal of Multivariate Analysis, Structural Equation Modeling, Psychological Methods, Psychometrika, British Journal of Mathematical and Statistical Psychology, Multivariate Behavioral Research等国际重要期刊上发表了上100篇学术论文。袁教授目前是Psychological Methods和 Journal of Multivariate Analysis两个期刊的副主编，并兼任多个期刊的编委.