# Political scientist bridges data science and social research validity Naoki Egami's methodology tackles AI biases and external validity in political studies. - **Source:** [MIT News — AI](https://news.mit.edu/topic/artificial-intelligence2) - **Published:** 2026-09-16 - **Topic:** [Data & AI](https://brifing.si/c/data-ai) - **Original article:** - **On Brifing:** https://brifing.si/s/7c47f44e-fc91-43f0-9a4b-b2969cbfe815 ## Challenges in social science research Social scientists face persistent questions about external validity—whether findings from one context apply elsewhere. Naoki Egami specializes in addressing these methodological gaps, combining political science with applied statistics. His work scrutinizes how variables like election competitiveness alter research outcomes. This approach helps standardize methodologies across diverse political environments. ## AI's role in research accuracy Egami began studying AI's impact on social science research before the ChatGPT boom. He examines how machine learning tools introduce systematic errors in data generation. His statistical methods account for these biases, ensuring replicability in studies using AI-generated data. This foresight positioned him as a pioneer in AI-augmented political methodology. ## From Tokyo to MIT Egami's career began when a chance workshop connected him with Kosuke Imai, now at Harvard. As a University of Tokyo math student interested in political philosophy, he found his calling through Imai's seminar on statistical social science. This led to Princeton University, where Imai became his PhD advisor, shaping Egami's focus on methodological rigor. ## Field experiments' hidden biases Egami identifies critical limitations in political field experiments. Campaigns often permit research only in non-competitive districts where outcomes seem certain. This creates skewed datasets that don't reflect battleground dynamics. His work helps researchers recognize how context affects voter behavior studies, improving experimental design. ## Award-winning interdisciplinary approach Egami received the 2025 Emerging Scholar Award from the Society for Political Methodology. His papers have won honors across three subfields: political methodology, experimental research, and political networks. This reflects his unique ability to connect statistical theory with practical political science challenges through adaptable methodologies. ## MIT's engineering mindset At MIT, Egami appreciates the 'spirit of engineering'—systematic problem-solving that complements his work. He teaches undergrad and graduate courses while continuing research on AI's research applications. The political science department's intense yet collaborative environment supports his interdisciplinary projects bridging data science and social science. ## Balancing theory and practice Egami emphasizes maintaining dual perspectives: solving technical statistical problems while addressing substantive political questions. He notes that pure theorists may overlook empirical challenges, while problem-focused researchers risk suboptimal solutions. This philosophy guides his four-year research cycles exploring new methodological frontiers. ## Future of political data science Egami's work establishes frameworks for next-generation political research amid AI proliferation. By developing error-aware statistical methods, he ensures machine learning enhances rather than compromises social science validity. His career demonstrates how methodological innovation can address democracy's evolving measurement challenges. ## Sources covering this story - https://news.mit.edu/2026/studying-society-accurately-naoki-egami-0916 — --- _Summary written by Brifing from the original reporting. Copyright in the original article belongs to its publisher; read the full piece at the link above._