Google recently released some mobility data. Skimming through the PDFs on the website shows you how much people have stopped moving around and started sticking to home. I wanted to know what that relationship looked like on the whole so I downloaded the PDFs and pulled the high-level numbers out for each country (picture below; EDIT: Just looked at the data and this is NOT one datapoint for each country. It’s one data point for each PDF which means each US State is also plotted.). The regression equation from this analysis:
Retail & Recreation % = -0.196 – 2.375 * Residential
To put it another way, even without any change in staying at home, retail and recreation activity are down 19.6%. Then for each additional 1% increase in staying at home, retail and recreation decreases by 2.375%.
The data goes to a lower geographic level for many of the files which is great because it shows significant heterogeneity in terms of response at the state and sometimes county level. It also comes with more missing data but I’m excited to see it when I get it all pulled together.
