报告人：袁克海   教授

时  间：12月26日（周一）上午9:00

地  点：经管学院335会议室

欢迎相关专业老师和研究生参加！  

报告简介：

In social, behavioral, education and health sciences, important attributes are often latent variables that cannot be observed directly, and have to be inferred from fallible measures. Their analyses are most effectively done by structural equation modeling (SEM). In contrast to statistical methods in traditional multivariate analysis, SEM has the mechanism of modeling manifest variables, latent variables, as well as measurement errors simultaneously. Although not familiar to many statisticians, SEM has become one of the most important methodology in many disciplines when analyzing survey or non- and quasi-experimental data.

I will give an review of SEM, including model formulation, parameter estimation, model evaluation, and its applications. Pros and cons of different methods will be noted, and the need for further development is pointed out.

报告人简介：

Education
1992-1995 Ph.D. in mathematics with concentration in statistics, UCLA
1985-1988 M.A. in statistics, Beijing Institute of Technology
1981-1985 B.S. in mathematics, Beijing Institute of Technology
Employment
2008- Professor, Department of Psychology
University of Notre Dame
2015 Spring Visiting Professor, Department of Psychology
The Chinese University of Hong Kong
2005-2008 O’Neill III Associate Professor, Department of Psychology
University of Notre Dame
2001-2005 Associate Professor, Department of Psychology & Lab for Social Research
University of Notre Dame
1998-2001 Assistant Professor, Department of Psychology
University of North Texas
1995-1998 Statistician, Department of Psychology, UCLA
1992-1995 Graduate Student Researcher, Department of Psychology, UCLA
1988-1992 Assistant Professor, Department of Applied Mathematics
Beijing Institute of Technology
Professional services
• Member of Editorial Board, Educational and Psychological Measurement (2000-).
• Member of Editorial Board, Structural Equation Modeling (2006-).
• Consulting Editor, Multivariate Behavioral Research (2006-).
• Member of Editorial Board, Journal of Educational and Behavioral Statistics (2011-).
• Member of Editorial Board, Educational Researcher (2013-).
• Member of Editorial Board, JSM Mathematics and Statistics (2014-).
• Member of Advisory Board, Behaviormetrika (2016-).
• Consulting Editor, Psychological Methods (2016-).
• Associate Editor, Journal of Multivariate Analysis (2008-2016).
• Associate Editor, Psychological Methods (2013-2015).
• Member of Editorial Board, Sociological Methodology (2007-2009).
• Reviewed grant proposals for NSF, Institute of Education Sciences, Spencer
Foundation, Research Council of Canada, The Research Grant Council of Hong
Kong, books for Lawrence Erlbaum Associates, Taylor and Francis Group, Guilford
Press, and manuscripts for over twenty journals.
Professional honors
• The James McKeen Cattell Sabbatical Award (2005).
• The Raymond B. Cattell Award for Early-Career Outstanding Multivariate Research
(2002) from the Society of Multivariate Experimental Psychology.
• Elected member of the Society of Multivariate Experimental Psychology (2002-).
Topics worked on
Mean comparison; regression; factor analysis; structural equation modeling; repeated
measures and multilevel modeling; mixture model; item response theory; mediation
and moderation analysis; post-hoc power, combining mean difference, asymptotics,
statistical computation; estimating equation; bootstrap and cross-validation;
nonnormal distribution; robust methods, missing data and small sample problems in
multivariate analysis.
Courses taught
Experimental psychology I: Statistics; Psychometric theory; Multivariate statistics;
Factor analysis; Structural equation modeling; Multilevel modeling; Computational
statistics; Statistical methods; Exploratory data analysis; Missing data analysis.
Doctoral dissertations directed
• Richard Herrington, University of North Texas (2001): Simulating statistical power
curves with the bootstrap and robust estimation.
• Ken Kelley, University of Notre Dame (2005, codirected with Scott Maxwell):
Estimating nonlinear change models in heterogeneous populations when class
membership is unknown: Defining and developing the latent classification differential
change model.
• Summer Zu, University of Notre Dame (2009): Robust procedures for mediation
analysis.
• Wei Zhang, University of Notre Dame (2010): Estimating latent variable interactions
with missing data.
• Xiaoling Zhong, University of Notre Dame (2010): Model selection, evaluation and
tests of invariance in finite factor mixture modeling using a two stage approach.
• Laura Lu, University of Notre Dame (2011, codirected with Zhiyong Zhang):
Bayesian inference of robust growth mixture models with non-ignorable missing data.
Publications/in press papers: (publications are classified by topics, each article
appears only in one category although it may touch more than one topic)