演讲人: Xin Tong,Quantitative Psychology, University of No

讲座时间: 6月26日（周3）下午：14:00-15:00

讲座地点: 信息楼0125室

讲座内容:

Structural equation modeling (SEM) is a popular multivariate modeling technique that is widely used in social, behavioral, and educational sciences. SEM allows for testing complex theories by modeling latent variables and measurement errors simultaneously. Many studies with real data have demonstrated that complete data with normal distributions are rather an exception, especially with data collected longitudinally. However, using traditional SEM techniques for such data may lead to inefficient or even incorrect parameter estimates. Therefore, many robust procedures have been proposed and developed. In this talk, several robust methods will be discussed and the application of these methods will be briefly illustrated through an example of the linear growth curve model.