Skip to main content

Understanding the Impact of COVID-19 Policy Change in the Greater Seattle Area using Mobility Data

Our take —

Reduced movement in the greater Seattle area has been observed, with more people staying at home, presumably in response to the broad array of social distancing measures progressively announced since March 5, 2020. Reduced movement does not necessarily equate to social distancing or a direct reduction in social contact. Other studies, including those from the same research group, consider the link between reduced mobility and SARS-CoV-2 transmission.

Study design


Study population and setting

Data from “Facebook Data for Good Project – Disaster Maps” were used to quantify and compare the movement of people in the Seattle metropolitan area from February 26, 2020 to March 28, 2020 relative to a baseline period of 45-90 days prior. Data were aggregated and measured movements among an average of 230,000 Facebook users per day who used mobile devices and enabled location services. In Washington State, beginning on March 4, 2020, state and local government agencies and private companies introduced a series of social distancing policies–working from home, bans of large gatherings, closure of recreational facilities and restaurants, school closures, and a shelter-in-place order. Observed changes in human movement were qualitatively compared with the timing of these social distancing measures.

Summary of Main Findings

There was a decline in overall movement within the region during the study period. As different policy measures were introduced, travel into Seattle from surrounding areas progressively decreased. Relative to the earlier period, movement into Seattle was 40% lower by March 5, 2020 (when major employers adopted work-from-home policies), 60% lower by March 12, 2020 (when large gatherings were banned and schools were closed), and 90% lower by the study’s end on March 28, 2020. Authors also observed less movement during the day, except for on some weekends. Daytime population occupancy counts increased in residential areas by an average of 27% relative to the earlier period.

Study Strengths

The authors noted high correlations between their measures of mobility from Facebook data and toll data from Washington State Traffic GeoPortal.


Mobility data were coarsely aggregated and identified neither individual movements nor social contacts. Movements of non-mobile Facebook users or those whose location services are not enabled were not captured. The resulting data may not be representative of either Facebook users or the greater Seattle-area population, though correlations with toll data are encouraging. The link between reduced movement and changes in social or physical contacts is unclear.

Value added

Findings are not novel, but confirm findings from other studies.