New technology plays a key role in political polarization around the globe. Recent years have witnessed an explosion of data from social media platforms and the digitization of massive governmental administrative records. The digital age has also created novel opportunities for online field experiments, new types of public opinion surveys, and mass collaboration. However, most social scientists do not yet have access to the technical training necessary to analyze the voluminous data generated through these digital environments each day. Scholars in the computer sciences and engineering possess the skills to process these data, but struggle to translate them into systematic theories of polarization or other human behaviors. Training the next generation of researchers at the intersection of social science and data science is crucial to advance the kind of research needed to address issues of public concern and inform policies and decision-making.

A team led by Chris Bail at Duke University’s Polarization Lab project plans to create 11 workshops to provide free training to more than 100 junior scholars within the nascent field of computational social science and incubate research across disciplinary lines. These polarization training sites would expand upon a pre-existing program, the Summer Institutes in Computational Social Science (SICSS), by focusing upon theoretical and methodological issues specific to the study of polarization and inter-group conflict.

All training materials produced at these sites will be added to SICSS’s open-source curriculum, and each site leader will be required to produce a public post-mortem of their event to inform future program sites and other training programs that may focus upon polarization and computational social science.