File talk:COVID-19 rolling 14day Prevalence in the United States by county.svg
Data error?
[edit]
The picture for Missuri seems rather off and does not seem to match other published data of infections with the last 7 days (see [1])--Kmhkmh (talk) 16:33, 16 March 2021 (UTC)
- @Kmhkmh: Actually, doesn't the link that you add above show exactly the same spike? [2] if you look at the NYT data that Google outlights? It's not visible here though. The NYT data does show a peak of "added antigen test positives" early march. It looks like the data is accurate, it just needs additional context to be understood. effeietsanders 19:19, 16 March 2021 (UTC)
- There is spike yes, but incidence value is 600 (for 7 days), whereas the map shows an incidence level > 1000 for much of the state. The other thing that high values being resticted to one state sticking out like a sore thumb without having any obvious difference in policies/measures/population to the surrounding states seems highly unlikely and probably more indicative of a problem with the (compiled) data (for whatever reasons). The Antigen test or new testing general might provide an explanation, though only under the assumption that they are exclusively rolled out in Missouri but nowhere else, which seems highly unlikely to as well--Kmhkmh (talk) 20:46, 16 March 2021 (UTC)
- @Kmhkmh: (please ping me if you want me to see your response) To my understanding, this is less about more tests being performed, and more about retroactively starting to count positive antigen tests. That is why the peak is only a single day. As you can see in the NYT link, that is about 6000 antigen positive cases added in a single day. I can't explain that other than through the assumption that they started to count some set of historically not counted tests, including all tests from the past year that were positive in that category. effeietsanders 02:30, 17 March 2021 (UTC)
- A week or two ago, Iowa was similarly the hotspot. I miss this map (I am looking for reported hotspots in Minnesota and Michigan). -SusanLesch (talk) 19:45, 19 March 2021 (UTC)
- @Kmhkmh: (please ping me if you want me to see your response) To my understanding, this is less about more tests being performed, and more about retroactively starting to count positive antigen tests. That is why the peak is only a single day. As you can see in the NYT link, that is about 6000 antigen positive cases added in a single day. I can't explain that other than through the assumption that they started to count some set of historically not counted tests, including all tests from the past year that were positive in that category. effeietsanders 02:30, 17 March 2021 (UTC)
- There is spike yes, but incidence value is 600 (for 7 days), whereas the map shows an incidence level > 1000 for much of the state. The other thing that high values being resticted to one state sticking out like a sore thumb without having any obvious difference in policies/measures/population to the surrounding states seems highly unlikely and probably more indicative of a problem with the (compiled) data (for whatever reasons). The Antigen test or new testing general might provide an explanation, though only under the assumption that they are exclusively rolled out in Missouri but nowhere else, which seems highly unlikely to as well--Kmhkmh (talk) 20:46, 16 March 2021 (UTC)
