The A(R)t and Science of Election Analytics with Soubhik Barari - nyhackr June Meetup

Talk Title- The A(R)t and Science of Election Analytics

Talk Description- This talk is a sampler of an important subdiscipline of data science that is, like it or not, here to stay and only growing: election analytics. From pre-election polling, to message experimentation; from opinion forecasting to voter turnout modelling, election analytics is built on tools and theories that span the data and behavioral sciences. In this talk, I will specifically discuss some of my own work in pre-election polling as a research scientist at SurveyMonkey during the 2022 midterms as well as an on-going project (in collaboration with Microsoft Research) to measure the coverage and effects of TV media in Congressional election campaigns. Along the way, I will showcase commonly used R tools in election analytics and I'll conclude with a broad discussion of future challenges and opportunities for election analysts, researchers, and practitioners.

Bio- Soubhik (sho-bik) is a computational social scientist and data scientist using quantitative methods to answer questions around media, political communication, public opinion, and behavioral science more broadly. Currently he is a Ph.D. student (soon-to-be graduate) in Political Science at Harvard University and a research scientist at SurveyMonkey. At Harvard he has co-taught courses on behavioral science, election analytics, and advanced statistical methods. Previously, he was a research intern at Microsoft Research and a staff data scientist at the MIT Political Methodology Lab. Soubhik holds an A.M. in Statistics from Harvard and a B.S. in Computer Science and Mathematics from Tufts.