My statistical research focuses on methods for correlated continuous and binary outcomes, which occur in infectious disease studies, policy evaluations, and survey research.

Theory and Applicability of Marginal Models for Correlated Binary Data

I am working on improving the understanding of intracluster correlation among binary data, especially in the presence of covariates. This work has led to methods that can improve sample size estimation for stratified cluster randomized trials. An article is forthcoming in Statistics in Medicine. The methods detailed there are available through a RShiny app at and in R code at Further research will enable more accurate sizing and power analysis for trials adjusted for individual-level covariates and observational studies with correlated data.

Analysis Methods for Infectious Disease Outbreak Trials

I have also conducted research on analysis methods for stepped wedge cluster randomized trials. These methods improve efficiency compared to fully non-parametric methods while being robust to a variety of data-generating settings. A preprint of this work is available on arXiv and the published article is now available from Statistics in Medicine. Future work will further clarify the properties of clinical trials in outbreak settings and identify the best designs for specific settings.

R code for these projects can be found at These projects are ongoing and the code is frequently updated.

History of Statistics and the Role of Statistics in Public Policy

Translational research is key to ensuring that statistics are properly used in fields as varied as public policy, business, law, and medicine, among others. I have written about the history of p-values and their role in public policy, especially at FDA. I will continue this work and the broader work of describing statistics to a wider audience. You can find links to these articles on the Writing page.

Fisher SMRW Table 2
R.A. Fisher, Statistical Methods for Research Workers, 1925. Harvard University Library.