As we’ve lived in and through this pandemic, employer/ employee decisions along with formal stay at home orders have led most of us to change where and when we go places, or if we even go to those places at all. We work from home (when and if we are able), shop less, and concerts, sports events, conferences, and other gatherings have been canceled. The strain and stress has been significant, but the social distancing involved in the changes has been crucial in slowing the spread of Coronavirus (COVID-19). Georgia’s statewide stay-at-home orders have been lifted, in fact, partly in response to the related changes in infection rates.

Using SafeGraph mobility tracking data, Atlanta Regional Commission developed a Social Distancing Dashboard to measure the effectiveness of social distance in controlling COVID-19. The data are collected through a panel sample of GPS “pings” from anonymous mobile devices. These pings can be analyzed to reveal the number of mobile devices that stayed at home, as well as the devices that were somewhere else (that had moved), at hourly scale. These data can then be used as a proxy to measure how people’s movements have decreased, or rather how social distancing has increased.

Click on the image below to access the dashboard. The mobile devices were grouped into three classifications: 1) staying at home devices, which did not leave their home location; 2) full-time behavior devices, which spent six (6) hours or more at a location other than home in the daytime (8 am – 6 pm); and 3) part-time behavior devices, which spent at least one period of 3 to 6 hours at a location other than home during the daytime (8 am – 6 pm).

The top section of the dashboard is a line graph showing the shares (of all devices) represented by each of the above groups, by week. A drop-down menu allows you to filter on individual Georgia counties. For counties in the 10-County Atlanta Regional Commission area, the share of  devices “staying at home“ increased steadily, from the week of March 9th on to a high for the week of April 6th. Clayton County had the least increase in percentage share of devices staying at home, while Fayette County had the largest percentage point increase. Let’s focus on the pattern in DeKalb County as an example.  There, for the week of March 9th, only about 25% of devices “stayed home”, but by the week of April 6th, this share had nearly doubled (in percentage point terms) to 47%. By the week of April 20, there was only a small downtick, to 45%, of the share of devices “staying at home”.

The second part of the dashboard (again, click on the image below for access) shows the weekly spatial pattern (at the census tract level) for the three device groupings, all the way from the first week of January to the week of April 20. You can use the slider button to select the week. Please note that when you select a county name from the legend in (upper) line graph portion of the dashboard, the county is then highlighted on the map series. If you select the week of March 2, you will see that high shares of devices “staying home” are concentrated solely in census tracts in the cores of metro areas. But then choose the week of March 16th, and you will note that most of the state (and almost all of metro counties) have higher shares of devices “staying at home”.

Data: SafeGraph