Seminar - Richard A. Levine

Event Date: 

Wednesday, October 28, 2020 - 3:30pm to 4:30pm

Event Location: 

  • Zoom Meeting

Title: Making an Impact in an Institutional Research Office: On Data Champions and Machine Learning


As a strategy to support data-informed decision making at SDSU, Analytic Studies & Institutional Research (IR) established a Statistical Modeling Group (SMG) within its operation. SMG is a collaborative team of machine learning experts from the Stat Dept and IR data management and visualization experts tasked with developing and applying predictive analytics methods to solve institutional effectiveness problems. We will highlight the role of SMG on our campus. Focusing on SMG success stories in STEM program retention and graduation success, we will 1) introduce the predictive analytics infrastructure and machine learning methods developed for student success efficacy studies; 2) show novel visualizations and dashboards developed for STEM advisors and campus administrators; 3) outline the Data Champions program instituted to expand University data capabilities and leverage our analytic tools to inform SDSU efforts to improve student success metrics; and 4) present our vision for Statistical Modeling Groups in IR units as an effective strategy for training Statistics/Data Science graduate students and delivering actionable information to campus stakeholders.
Rich Levine is Professor of Statistics and Faculty Advisor for Analytic Studies and Institutional Research at San Diego State University. He also served as Chair of the SDSU Mathematics and Statistics Department. Rich is an ASA Fellow and currently Associate Editor of Statistics for the Notices of the American Mathematical Society. He previously served as Editor of the Journal of Computational and Graphical Statistics, overall Program Chair for the 2019 Joint Statistical Meeting in Denver, CO, and Fulbright Scholar to Zhejiang University School of Medicine in Hangzhou, China. His recent research interests entail developing and applying machine learning methods in educational data mining settings. In a past research life he worked on MCMC and Bayesian computing problems. He enjoys thinking about how to incorporate statistical literacy and statistics communication into data science curricula, a component of which he will share in this talk.
Richard Levine