Friendly Cities is currently working with:
- Dr. Dima Nazzal and Dr. Lauren Steimle of GaTech ISyE on interactive maps for the campus re-opening plan.
- The Weitz Group (led by Dr. Joshua Weitz) to produce research and an interactive map of COVID-19 risk by county.
- Thanks to graduate students Seolha Lee and Yilun Zha for their map designs on both projects.
Online technical report: Movement and COVID-19
Lee, S., Liang, X., and Andris, C. (2020), “COVID-19 Behavioral Responses.” Friendly Cities Lab Technical Report, Georgia Institute of Technology. Accessed online: friendlycities.gatech.edu/covid19.
QUESTION 1: What kinds of jobs/industries are associated with more movement or less movement? Counties with extraction, manufacturing, utilities, business management, agriculture, retail trade, transportation and warehousing, and wholesale trade have reduced their mobility the least. Counties with high levels of finance and insurance, professional, scientific and technical employment have reduced their mobility the most.
Mobility data from: Descartes Labs ‘M50’ mobility index, representing the change in the median distance people in a county are traveling (top 10% of anomalies removed). Job data from: county-level NAICS codes (types of jobs people have). Counties sectioned by top 25% and bottom 25% quantile; Top counties (n = 502) reduced their mobility <17% of typical mobility rates and bottom counties (n = 501) reduced mobility >35% of typical rates. We only account for counties with shelter-in-place (by 4/3/20). Mobility data accounts for weekdays from the start of the order to 4/10 so all counties can react to the order. Results are not sensitive to the percentage of residents working outside their home county, which is 50% for both groups via commuter data. Below is a map defining the counties used.
Question 2: Which parts of the world are staying home & avoiding non-essential retail+entertainment? European and South American countries are, on the whole, staying home and avoiding unnecessary trips. Malaysia’s strict orders are paying off and Bolivia, Sri Lanka and Panama are reducing their trips. New Zealand’s winning approach and Japan’s emerging crisis can also be seen. U.S. states vary in response rates, with Northeastern states changing their behavior and Southern states neglecting to do so.
Plot captures the situation on 3/29/20. Trip data is from Google’s PDFs (inc. retail, grocery stores, parks, home, jobs, etc.) via Vitor Baptista. Deaths are from USAFacts and European Centre for Disease Prevention and Control (ECDC). (Download the GIS data at US county level here).
Question 3: Which counties have sufficient ventilators and hospital beds per capita?