This specialization focuses on applications of public opinion analytics and data science in political science. Students are encouraged—not required—to have basic preparation in political science. Prior completion of “Political Science I & II” (6 credits, departmental core) or “Introduction to Political Science” (2-credit general education) is recommended to ensure fluency with key concepts and debates.<br>Level 1 — Programming for the Social Sciences & An Introduction to Statistical Learning<br>Begin with a practical toolkit. In Programming for the Social Sciences, students learn to operationalize computational methods for political communication and social issues. The emphasis falls on working with text?as?data, network structures, demographic variables, and geographic information—building habits for handling data types commonly encountered in social research and for acquiring large-scale digital traces. In Introduction to Statistical Learning, students adopt a “learn by doing” approach to applied inference. Using R, they practice data wrangling, exploratory analysis, and core estimation techniques, gaining a conceptual grasp of statistical learning while developing reproducible workflows.<br>Level 2 — Special Topics in Public Opinion Analytics and Election Studies<br>Next, students connect theory to measurement. The course introduces major theories of public opinion and voting behavior and surveys research designs used in election studies. From the voter’s perspective, students analyze assumptions about political cognition and the determinants of vote choice, including socioeconomic background, group identification (e.g., party identification, class voting, ethnic attachments), issue orientations, candidate evaluations, social networks, and phenomena such as strategic and split-ticket voting. From the institutional perspective, students examine how electoral rules are designed and how those rules shape party systems, competition, and representation. By the end, students can navigate diverse election datasets and select appropriate methods for answering substantive questions.<br>Level 3 — Computational Social Science<br>Students then broaden their methodological range. This course introduces the core tools and techniques of computational social science and builds them around concrete research questions. Drawing on mathematics, statistics, computer science, and data science, students engage problems in political science, sociology, economics, management, and allied fields. They learn to implement advanced computational approaches—including mathematical modeling, social network analysis, and complexity-informed theories—to study collective behavior, information diffusion, and institutional dynamics at scale.<br>Level 4 — Causal Inference for the Social Sciences<br>Finally, students consolidate their ability to make credible causal claims. Building on experience with large, complex datasets, the course develops strategies for uncovering cause-and-effect relationships using observational data, statistical methods, and experimental designs. Students examine the logic of identification, study designs suited to policy-relevant questions, and the trade-offs among randomized experiments, quasi-experiments, and observational estimators. Through guided applications, they design and implement analyses that move from descriptive patterns to causal explanations for events of substantive and policy significance.