SUBMITTED/E-PRINTS

Mozer, R., Kaufman, A. R., Celi, L. A., & Miratrix, L. Leveraging text data for causal inference using electronic health records. (Under Review) arXiv preprint arXiv:2307.03687

Kim, Y., Mozer, R., Miratrix, L., and Al-Adeimi, S. ChatGPT vs. Linear Regression: Assessing the Efficacy and Accuracy of Large Language Models for Automated Essay Scoring.

Published

Rupcic S, Tam MZ, DeLaughter KL, Gifford AL, Barker AM, Bokhour BG, Xu C, Dryden E, Anderson E, Jasuja GK, Boudreau J, Douglas JH, Hyde J, Mozer R., Zeliadt SB, Fix GM. (In press) Co-Designing a Blueprint for Spreading Person-Centered, Whole Health Care to HIV Specialty Care Settings: A Mixed Methods Protocol. BMC Health Services Research.

Mozer, R. & Miratrix, L. (In press) More power to you: Using machine learning to augment human coding for more efficient inference in text-based randomized trials. Annals of Applied Statistics.

Mozer, R., Miratrix, L. Relyea, J. E., & Kim, J. S. (2024) Combining human and machine scoring in experimental assessments of writing: a case study tutorial. Journal of Educational & Behavioral Statistics, 49(5), 780-816.

Mozer, R. & Glickman, M.E. (2023). Bayesian analysis of longitudinal studies with treatment by indication." Health Services & Outcomes Research Methodology, 23(4), 468-491.

Anderson, E., Dvorin, K., Etingen, B., Barker, A. M., Rai, Z., Herbst, A., Mozer, R., Kingston, R. P., and Bokhour, B. (2022). ‘It Makes You Sit Back and Think Where You Wanna Go’: Veteran experiences in virtual whole health peer-led groups. Health Expectations, 25, 2548-2556. doi:10.1111/hex.13581

Rahman, N., Mozer, R., McHugh, K., Rockett, I., Chow, C., & Vaughan, G. (2022) Using natural language processing to improve suicide classification requires consideration of race. Suicide and Life Threatening Behavior, 52, 782– 791. https://doi.org/10.1111/sltb.12862.

Anderson, E., Dvorin, K., Etingen, B., Barker, A. M., Rai, Z., Herbst, A., Mozer, R., Kingston, R. P., and Bokhour, B. (2021). Lessons learned from VHAs rapid implementation of virtual whole health peer-led groups during the COVID-19 pandemic: Staff perspectives. Global Advances in Health and Medicine, 11, 1-12.

Mozer, R., Miratrix, L., Kaufman, A. R., & Anastasopoulos, L. J. (2020). Matching with text data: An experimental evaluation of methods for matching documents and of measuring match quality. Political Analysis, 28(4), 445-468.

Mozer, R., Rubin, D., & Zubizarreta, J. (2020). Statistical Inference for Causal Effects in Clinical Psychology: Fundamental Concepts and Analytical Approaches. In A. Wright & M. Hallquist (Eds.), The Cambridge Handbook of Research Methods in Clinical Psychology (Cambridge Handbooks in Psychology, pp. 415-425). Cambridge: Cambridge University Press. doi:10.1017/9781316995808.038

Kessels, R., Mozer, R., & Bloemers, J. (2019). Methods for assessing and controlling placebo effects. Statistical Methods in Medical Research, 28(4), 1141-1156.

Bavli, H. J., & Mozer, R. (2018). The effects of comparable-case guidance on awards for pain and suffering and punitive damages: Evidence from a randomized controlled trial. Yale L. & Pol'y Rev., 37, 405.

Mozer, R., Kessels, R., & Rubin, D. B. (2017). Disentangling treatment and placebo effects in randomized experiments using principal stratification—an introduction. In The Annual Meeting of the Psychometric Society (pp. 11-23). Springer, Cham.


In Progress

Guo, G., Branson, Z., & Mozer, R. Doubly robust causal inference with text-based confounding: An application to gender disparities in online forums. Working paper.

Mozer, R.
& Mealli, F. Causal Inference with Complex Data: A Guide for the Modern Statistician.


OTHER

Reagan Mozer and Donald B. Rubin. "Evaluating PML Risk for Multiple Sclerosis Patients Receiving Tysabri". Manuscript upon request.

Brenda Osuna and Reagan Rose (2013). "Applying Quantitative Content Analysis and Factor Scoring: Comparison of Pre- and Post-9/11 Song Lyrics." In JSM Proceedings, Statistical Computing Section.


SOFTWARE

textmatchR package for matching text documents and performing quality/balance assessments in studies where covariates and/or outcomes are defined by aspects of text
(under development, beta version available at: https://github.com/reaganmozer/textmatch)

Replication materials for: “Matching with Text Data: An Experimental Evaluation of Methods for Match- ing Documents and of Measuring Match Quality”. Published on the Political Analysis Dataverse hosted by Harvard Dataverse, doi:10.7910/DVN/K8IL3V