Research Interests
My research activities center primarily on the development and application of hierarchical modeling techniques. This work essentially falls into three categories:
1. The use of hierarchical models (HMs) in multisite evaluation studies: My methodological work in this area focuses on showing how HMs can be used to investigate the conditions under which educational programs are successful, and for whom. I am also interested in techniques for studying the sensitivity of inferences in multisite evaluations to unmeasured potential confounding variables. In terms of substantive applications, I have collaborated on multisite studies of innovative mathematics curricula and remedial reading programs.
2. The use of HMs in studies of change: One strand of my work this area centers on the use of piecewise growth models in studying patterns of individual change across qualitatively distinct periods of time (e.g., the treatment and follow-up phases of an intervention study). A second strand of work focuses on the estimation and use of growth models in which initial status is employed as a predictor of rates of change. With regard to applications of growth modeling techniques, I have collaborated on longitudinal studies of short-term psychotherapy interventions and studies of language development.
3. Bayesian estimation of HMs using Markov chain Monte Carlo (MCMC) techniques: My work in this area centers on using MCMC to obtain robust estimates of parameters of interest in HMs. This entails fitting HMs under t distributional assumptions. My work on employing initial status as a predictor of rates of change (see 2. above) also involves the use of MCMC.
I am also interested in exploring the philosophical underpinnings of inquiry, in particular key epistemological and ontological issues that cut across the qualitative / quantitative divide (e.g., the theory-laden-ness of observation; the value of attending to alternative explanations).
Publications within the last five years
- Seltzer, M. & Rose, M. (in press) Constructing analyses: Developing thoughtfulness in working with quantitative methods. In C. Conrad & R. Serlin (Eds.), Handbook for Research in Education: Engaging Ideas and Enriching Inquiry. Thousand Oaks, CA: Sage Publications.
- Kaplan, D. (Ed.) (Section Editors: R. Hoyle, R. Nandakumar, S. Mulaik, S. Nishisato, D. Rindskopf & M. Seltzer) (2004). The Handbook of Quantitative Methods for the Social Sciences. Thousand Oaks, CA: Sage Publications. (I am the editor of the section of the handbook that focuses on multilevel models (pp. 215-297).)
- Seltzer, M. (2004). The use of hierarchical models in analyzing data from experiments and quasi-experiments conducted in field settings. In D. Kaplan (Ed.), The Handbook of Quantitative Methods for the Social Sciences (pp. 259-280). Thousand Oaks, CA: Sage Publications.
- Svartberg, M., Stiles, T. & Seltzer, M. (2004). Effectiveness of short-term dynamic psychotherapy and cognitive therapy for Cluster C personality disorders: A randomized controlled trial. American Journal of Psychiatry, 161, 810-817.
- Seltzer, M Seltzer, M., Choi, K. & Thum, Y.M. (2003). Examining relationships between where students start and how rapidly they progress: Using new developments in growth modeling to gain insight into the distribution of achievement within schools. Educational Evaluation and Policy Analysis, 25, 263-286.
- Seltzer, M. & Choi, K. (2002). Model checking and sensitivity analysis for multilevel models. In S. Reise & N. Duan (Eds.), Multilevel modeling: Methodological advances, issues, and applications. Hillsdale, NJ: Lawrence Earlbaum.
- ., Novak, J. Choi, K. & Lim, N. (2002). Sensitivity analysis for hierarchical models employing t level-1 assumptions. Journal of Educational and Behavioral Statistics, 27, 181-222.
Recent courses
Education 230A: Introduction to Research Design and Statistics
Education 231D: Hierarchical Linear Models
Education 255C: Advanced Hierarchical Modeling
Education 243: Reflections on Social Science Research Methods
