Postdoctoral 
Research Associate, The Methodology Center
The Methodology Center
The Pennsylvania State University
400 Calder Square II
State College, PA 16801
814-867-4504
Education
Ph.D. University of California at Davis, 2010 (Quantitative Psychology)
M.A. University of California at Davis, 2007 (Quantitative Psychology)
B.S. East China Normal University, Shanghai, P.R. of China, 1998 (Psychology)
Research Interests
My primary quantitative research interests include multilevel models, structural equation models, missing data methods, longitudinal data analyses and Bayesian methods. I am also interested in the application of these methods in prevention studies.
Current Projects and Collaborators
I am currently working with Donna Coffman on causal mediation models. I am working on implementing propensity score methods in multivariate data analysis for substance use and HIV research. I also work with Stephanie Lanza on the software development and application of advanced latent class modeling techniques. My work involves testing and documenting the use of LCA procedures.
Honors and Awards
- 2006–2008: UC Davis Graduate Student Block Grant
- 2005–2006: Pacific Rim Research Program Mini-grant, University of California, Office of the President
- 2004–2005: UC Davis Graduate Student Fellowship
Publications
Peer-Reviewed Articles & Book Chapters
Xu, S., & Blozis, S. A. (2011). Sensitivity analysis of mixed models for incomplete longitudinal data. Journal of Educational and Behavioral Statistics 36(2), 237-256.
Blozis, S. A., Cho, Y.I., & Xu, S. (2010). Latent curve models: A structural equation perspective. Sociological Methods & Research, 39, 297-304. (An invited book review)
Xu, S., Blozis, S.A., & Ge, X. (2007). Missing data techniques for family studies: A comparison of listwise deletion, full information maximum likelihood, and multiple imputation. Early Growth and Development Study (EGADS) Tech Report.
Li, G. Y., & Xu, S. (1999). A study on the relationship between adaptive ability and home environment in middle school. Huaxia Press, 45-58.
Cui, L.J., Xu, S., & Wang, X. H. (1998). The belief and modeling of aging. Shanghai Science on Aging, 20, 55-58.
Presentations
Lanza, S. T., Coffman, D. L., Xu, S., & Almirall, D. (2011, June). Causal inference in latent class analysis. In B. C. Bray (Chair), Innovations in latent class analysis: New approaches to address classic questions. Symposium presented at the Society for Prevention Research Annual Meetings, Washington, DC.
Xu, S., Blozis, S., & Coffman, D. L. (2011, June). Alternative models for longitudinal variables with many zeros. Poster presented at the Society for Prevention Research Annual Meetings, Washington, DC.
Dziak, J. J., Huang, L., Lanza, S. T., Li, R., Collins, L. M., & Xu, S. (2011, June). Software advances from the Methodology Center at Penn State. Technology demonstration presented at the Society for Prevention Research Annual Meetings, Washington, DC.
Xu, S., & Blozis, S. A. (2009). Analysis of longitudinal data with many zeros: A comparison of bayesian and maximum likelihood procedures. Paper presented at the American Psychological Association Annual Meeting, Toronto, Canada.
Xu, S., & Blozis, S. A. (2008). A two-part mixed-effects model for time-use data: A comparison of maximum likelihood and bayesian estimation. Paper presented at the International Meeting of the Psychometric Society, Durham, N.H.
Xu, S., Blozis, S. A., & Conger, K. J. (2006, March). Tobacco smoking from adolescence to early adulthood: A longitudinal investigation. Poster presented at Society for Research on Adolescence, San Francisco, CA.