search

CSSEGISandData / COVID-19

Novel Coronavirus (COVID-19) Cases, provided by JHU CSSE
https://systems.jhu.edu/research/public-health/ncov/
29.1k stars 18.38k forks source link

Testing data revision #3042

Open RuijingZ opened 4 years ago

RuijingZ commented 4 years ago

Hello,

Thank you for preparing all these data.

I aware that you use COVID tracking project data for testing data. I see they changed methodology on how to count testing data by states. They basically switch from using calculated positive+negative totals to using explicitly reported total tests figures. Therefore, all historical data have been changed back to March. Here is the detailed explanation: https://covidtracking.com/blog/counting-covid-19-tests

I would I like to ask if you will apply these revisions to historical data as well?

Best, Ruijing

NaKo91 commented 4 years ago

I'd like to take the opportunity to point to the data you use for calculating the positivity rate of testing. You take the numbers of conducted tests from The Covid Tracking Project, but instead of using your own data of daily confirmed cases, you use their data for this item as well. As their numbers differ from yours constantly, I wonder why you don't rely on your own case count for calculating the positivity rate.