Can we learn anything from the Google mobility data
Since the middle of February 2020, Google has been making data available to the public on movement trends by geography across different categories of places such as retail and recreation, groceries and pharmacies, parks, transit stations, workplaces and residential. The reports shows the percentage increase or decrease from a base line measured over the 5-week period 3 January to 6 February 2020.
For residential, the base line measurement is the number of hours spent at home. If we assume a base line measurement of 16 hours a day spent at home (with 8 hours at work, travel or recreational), then an increase of 35% in the Google mobility data would imply that people were spending 22 hours a day at home (16 hours x 1.35).
The graph below plots the residential mobility data supplied by Google for each of the nine provinces of South Africa. The data shown is the daily average for each of the 30 weeks provided in the latest Google report.
The time spent at home increased by 30% to 35% during the first phase of lockdown and this percentage has since come down in steps in line with the easing of the lockdown restrictions. In the first week of September the average person is still spending an estimated additional 1.6 hours a day at home (16 hours x 10%) or more likely a good proportion of us are working from home while the balance have resumed normal work patterns.
It is interesting to compare the mobility data with the NMG forecast for the numbers infected with the SARS-CoV-2 virus as shown in the graph below. The graph expresses the numbers infected as a percentage of the population in the province.
On the face of it lockdown would appear to have delayed the peaks in the viral infections in each of the provinces. For the Western Cape, one can even advance an argument that lockdown has 'flattened the curve' of viral infections.
We continue to update the NMG model on the pandemic in South Africa and the latest model is available at https://nmg-covid-19.sctechnology.co.za/. I will also be monitoring the Google mobility data to see if and when the patterns go back to pre-COVID levels.
17 September 2020
Comments
Post a Comment