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 the modeling of student growth. Recent interests include multidimensional item response theory (MIRT) applications to novel item formats such as in computer-based assessments. Feel free to contact me for information about the Quantitative Methods program area.
My interests are in the quantification of uncertainty through the framework of Bayesian statistics. I am especially interested in Bayesian approaches to model averaging, missing data, probabilistic forecasting, and dynamic borrowing from historical data. My work is applied to problems in the design and analysis of large-scale educational assessments such as the Program for International Student Assessment (PISA) and the National Assessment for Educational Progress (NAEP). My research website can be found here.
My research interests are concerned with the development and application of quantitative methods in the social and behavioral sciences. I am particularly interested in causal inference with experimental, quasi-experimental, and observational data, multilevel models for clustered data, continuous and discrete growth models for examining patterns of change in human development with longitudinal data, and various statistical methods for investigating individual differences and heterogeneous treatment effects, such as latent variable models and machine learning approaches.
My research involves developing statistical methods for problems in education, psychology, and other areas of social science research, with a focus on methods related to research synthesis and meta-analysis. Meta-analysis is an area of statistical methodology concerned with how to combine evidence drawn from multiple sources, such as multiple primary studies on a closely related topic. Within meta-analysis, I focus on developing methods and tools for handling the complex data structures that commonly occur in application, such as by using robust variance estimation methods. I have worked extensively on methods for synthesizing data from single-case experimental designs. I am also interested in methods for detecting and mitigating the problems created by publication bias and outcome reporting bias. I collaborate regularly with researchers conducting large-scale evidence syntheses. Beyond meta-analysis, I am interested in statistical methods for causal inference in experimental and quasi-experimental contexts, robust methods of statistical inference, and open source software development, especially in R.
My interests are in test construction, test administration, and item response theory. Broadly speaking, my research program has focused on applications of item response theory to improve the validity and interpretability of test scores. More specifically, my scholarly activity has centered predominantly on two aspects of test security: The first area is developing and refining statistical methods to detect test fraud, including detection of group-based cheating (such as preknowledge, collusion, and answer changing), detection of individual cheating (such as answer copying), and detection of item compromise. The second area is defining and improving industry policy and best practices with regard to test security, including approaches implemented by testing programs to prevent, deter, impede, and detect cheating on tests.