This seminar explores the innovative use of SafeGraph mobility data to understand the patterns and implications of school closures during the COVID-19 pandemic. The speaker introduces their research methodology, which combines geolocation data from mobile devices with publicly available school district data to estimate the likelihood that a school was open for in-person instruction at any given time. The data spans various phases of the pandemic, offering a dynamic picture of how closures evolved over time and space.
A major theme of the talk is the pronounced disparity in access to in-person education. The researcher demonstrates that school closures were not randomly distributed. Instead, they were strongly associated with both political and socioeconomic factors. For instance, schools in wealthier, suburban, and Republican-leaning areas were more likely to reopen earlier, while those in lower-income, urban, and Democratic-leaning areas tended to remain closed longer. These findings reveal a concerning pattern of educational inequality exacerbated by the pandemic response.
The researcher also discusses the technical challenges of using mobility data to infer school status, such as distinguishing between students and staff or between schools and other types of locations. Despite these challenges, the team validated their approach by comparing it with school district announcements and survey-based data. This methodological rigor lends credibility to their insights and shows the power of combining commercial mobility datasets with traditional public records for academic research.
In the concluding portion of the seminar, the speaker highlights the broader implications of their findings. Beyond documenting historical patterns, their work provides a framework for future real-time monitoring of educational access during crises. It also raises important ethical and policy questions about data privacy, equity, and the role of data science in shaping public health and education policy. The presentation is a compelling example of how novel data sources can help illuminate social inequalities during times of upheaval.