Richard Berk

Professor of Criminology and Statistics

Department of Criminology and Department of Statistics

Ph.D., Sociology, The Johns Hopkins University, 1970

B.A., Psychology, Yale University, 1964

Professor Berk has appointments in the Department of Criminology and the Department of Statistics. He works on various topics in applied statistics including causal inference, statistical/machine learning, and methods for evaluating social programs. Among his criminology applications are inmate classification and placement systems, law enforcement strategies for reducing intimate partner violence, detecting violations of environmental or worker safety regulations, claims that the death penalty serves as a general deterrent, and forecasts of criminal behavior and/or victimization using statistical/machine learning procedures.

Some Recent Peer-Reviewed Publications:

“Valid Post-Selection Inference,” (with Lawrence Brown, Andreas Buja, Edward George, Kai Zhang, and Linda Zhao), Annals of Statistics 41(2): 802-837, 2013.

"Misspecified Mean Function Regression: Making Good Use of Regression Models that are Wrong." (with Lawrence Brown, Andreas Buja, Edward George, Emil Pitkin, Kai Zhang, and Linda Zhao), Sociological Methods and Research, 43: 433-451, 2014.

"Forecasting Domestic Violence: A Machine Learning Approach to Help Inform Arraignment Decisions," (with Susan B. Sorenson and Geoffrey Barnes), Journal of Empirical Legal Studies," 31(1): 94-115, 2016. 

"An impact assessment of machine learning risk forecasts on parole board decisions and recidivism” Journal of Experimental Criminology 13(2) 193-216, 2017

“Fairness in Criminal Justice Risk Assessments: The State of the Art,” (with Hoda Heidari, Shahin Jabbari, Michael Kerns, and Aaron Roth), Sociological Methods and Research, first published online July 2nd, 2018: http://journals.sagepub.com/doi/10.1177/0049124118782533

"Modern Neural Networks Generalize Well on Small Datasets,” (with Matthew Olson, Abraham Wyner), Conference Proceedings, 32nd Conference on Neural Information Processing Systems (NIPS 2018, Montréal, Canada)

"Machine Learning Risk Assessments in Criminal Justice Settings," New York: Springer, 2018.

“Assumption Lean Regression,” (with Andreas Buja, Lawrence Brown, Edward George, Arun Kumar Kuchibhotla, and Linda Zhao), The American Statistician, 2019. DOI:10.1080/00031305.2019.1592781

“Models as Approximations, Part I: A Conspiracy of Random Regressors and Model Deviations Against Classical Inference in Regression,” (with Andreas Buja, Lawrence Brown, Edward George, Emil Pitkin, Mikhail Traskin, Linda Zhao, and Kai Zhang), Statistical Science, forthcoming, 2019.

“Models as Approximations, Part II: A General Theory of Model-Robust Regression,” (with Andreas Buja, Lawrence Brown, Ed George, Arun Kumar Kuchibhotla, and Linda Zhao), Statistical Science, forthcoming, 2019.

“Using Recursive Partitioning to Find and Estimate Heterogenous Treatment Effects In Randomized Clinical Trials”  (with Matthew Olson, Andreas Buja, and Aurelie Ouss),  Journal of Experimental Criminology, forthcoming, 2020.

“An algorithmic Approach to Forecasting Rare Violent Events: An Illustration Based on Intimate Partner Violence Perpetration,” (with Susan B. Sorenson), Criminology and Public Policy, forthcoming, 2020.

 

 

 Professional Awards Elected to the Sociological Research Association; Elected Fellow to the American Association for the Advancement of Science; Paul S. Lazarsfeld Award for methodological contributions from the American Sociological Association; Elected Fellow of the American Statistical Association; Elected Fellow of the Academy of Experimental Criminology.

https://crim.sas.upenn.edu/sites/default/files/BerkCV.pdf