University of Wisconsin–Madison

Quantitative Methods Faculty Research

Daniel Bolt, Professor

My interests are in the theory and application of psychometric methods in education and psychology. I am especially interested in the application of latent variable models for purposes of test validation, assessment of individual differences (such as response styles), and cognitive diagnosis. Most of my research is in item response theory (IRT), including its application to issues such as differential item functioning and test dimensionality assessment. I am also interested in the development of nonparametric IRT methods, which relax certain modeling assumptions and have the potential to increase the flexibility and efficiency of IRT in many testing applications.

Jee-Seon Kim, Associate Professor

My research interests are concerned with the development and application of quantitative methods in the social and behavioral sciences. I am particularly interested in multilevel models and other latent variable models, methods for modeling change, learning, and human development using longitudinal data, categorical data analysis, and issues related to omitted variables. I have explored various applications of these methods for studying individual differences, patterns of change, and school effectiveness. I enjoy connecting theoretical models to real-world problems and have sought to use methodology to address practical issues in the behavioral sciences. I find this interdisciplinary aspect of my work to be rewarding and exciting, as it provides a strong sense of purpose to my ongoing research program.

David Kaplan, Professor

My current program of research focuses on the development and testing of statistical models for social and behavioral processes that are not necessarily directly observed. Latent variable models, growth curve models, mixture models, and Markov models can be used to study unobserved processes and together constitute statistical methodologies that interest me. I am currently most interested in the application of Bayesian inferential methods applied to these methodologies.

I am also interested Bayesian approaches to causal inference in experimental and quasi-experimental settings. In the experimental setting, I am particularly interested in the use of Bayesian informative hypotheses to guide the testing of causal claims. In the quasi-experimental setting, my current focus of research is on applications and developments in Bayesian propensity score modeling and Bayesian model averaging as a means of improving causal estimands.

My collaborative research involves applications of advanced quantitative methodologies to problems in educational psychology, human development, and international comparative education. I am most actively involved in the OECD Program for International Student Assessment (PISA) where I serve on its technical advisory group and questionnaire expert group.

Peter Steiner, Assistant Professor

My current program of research focuses on the methodology of causal inference, particularly quasi-experimental designs in education. These designs include non-equivalent control group designs, regression discontinuity designs, and interrupted time series analysis. I am currently most interested in the theory and practice of propensity score techniques for matching non-equivalent control groups, particularly under which conditions propensity score methods work in actual research practice. I am also interested in the design and analysis of factorial surveys, also called vignette analysis, where respondents are typically confronted with a set of factorially varied descriptions of situations or persons.

James Wollack, Associate Professor

My research program has predominantly focused on applications of item response theory to improve the validity and interpretability of test scores. Much of my research has centered on aspects of test security, and most specifically, detection of cheating, including developing indexes for detecting answer copying, studying the properties of different indexes under various testing conditions, helping establish best practices for implementing cheating indexes, and identifying strategies for improving detection rates of true copiers. I am also interested in studying the effects of violating assumptions underlying common measurement models (e.g., item parameter drift, item misfit, and local item dependence) and developing approaches to mitigate these effects.