Sara Ji is a doctoral candidate at Harvard University whose work sits at the intersection of the economics of education, labor economics, and causal inference. Her research examines how school-based experiences and institutional policies shape long-run educational, labor market, and civic outcomes, using large-scale administrative data and newly constructed micro-level datasets. A central strand of her work studies student mobility, particularly the Metro Council of Education Opportunity (METCO) voluntary desegregation program. She builds linked administrative datasets spanning applications, school assignments, and long-run outcomes such as educational attainment and civic participation. Her work leverages quasi-experimental variation from program assignment rules, timing, and capacity constraints to study the effects of access to higher-opportunity school environments, as well as broader intergenerational consequences of educational mobility.   

Another major project studies how access to extracurricular activities in high schools affects student outcomes, with a focus on equity-relevant policy environments such as Title IX. Sara constructs a novel dataset on extracurricular offerings in Massachusetts high schools using yearbooks and administrative records, applying machine learning and text extraction methods to measure clubs, sports, and student organizations at scale. The project examines how institutional rules, school resources, and gender equity enforcement shape the supply of extracurricular opportunities, and how changes in access to these activities relate to academic engagement, postsecondary trajectories, and longer-run outcomes.  

Sara also works on applied policy evaluation projects in education, including studies of academic mentoring programs, curriculum dissemination, and teacher recognition systems. These projects typically use lottery-based designs, difference-in-differences, and event-study frameworks to estimate impacts on student achievement, attendance, graduation, and college enrollment. Methodologically, her work emphasizes careful data engineering, linking across administrative sources, and transparent causal identification strategies. She frequently works with longitudinal education datasets, census-linked population measures, and administrative records, and is developing a research agenda that connects school inputs to long-run social and economic outcomes.