South Dakota

Dashboard: https://app.powerbigov.us/view?r=eyJrIjoiNTNlZGFlNzgtYmJjNC00ZjZlLTg4MDgtYmNkNTdiY2MxMDUyIiwidCI6IjcwYWY1NDdjLTY5YWItNDE2ZC1iNGE2LTU0M2I1Y2U1MmI5OSJ9 Tab: Trends Chart: PCR Test Positivity Rate by Date Reported to SD-DOH

Dashboard: https://app.powerbigov.us/view?r=eyJrIjoiNTNlZGFlNzgtYmJjNC00ZjZlLTg4MDgtYmNkNTdiY2MxMDUyIiwidCI6IjcwYWY1NDdjLTY5YWItNDE2ZC1iNGE2LTU0M2I1Y2U1MmI5OSJ9 Tab: Trends Chart: PCR Test Positivity Rate by Date Reported to SD-DOH

State or US: SD

Describe the problem Total tests was entered as 53622, needs to be 52622, according to pos+neg

Link to data source https://covidtracking.com/screenshots/SD/SD-20200604-184435.png

[State]: South Dakota [Dates impacted]: 8/25-8/26 [Issue described]: On 8/27 SD's update included cases that were not reported on August 25th and 26th due to a reporting aberration. They also included the corrected data for cumulative cases and cumulative people tested, which we can backfill.

State or US: South Dakota

Date Range: 3/15 - 5/13 Data Points: Total PCR Tests (People)

Describe the problem On 8/3, SD updated their dashboard to include "Total Tests," which we have recorded in Total Tests (PCR).

We also have Total PCR Tests (People), but this only goes back until 5/13. After some digging, I found that all of our Total PCR Tests (People) values are sums of the pos+neg values from the pie chart on the SD dashboard. This pos+neg pie chart dates back until 4/20, and prior to this, they have a table separating pos/neg through our earliest screenshot for 3/15. We should be able to sum these pos+neg values for a complete time series.

The 3/15 screenshot shows data "as of" 3/14 and confirms that these values refer to individual people:

Link to data source State screenshots

Describe the Issue: On 8/2 it was discovered that SD has changed their data as of time to 13:00, we are now using previous day 13:00 as the timestamp. Before this change we were using previous day 17:00. This change was made on 7/13, so we should patch timestamps from 8/1 - 7/13 using previous day 13:00.

Data Source: https://doh.sd.gov/news/Coronavirus.aspx

State: SD

Issue description: We can't confirm that antibody testing isn't lumped in with RT-PCR testing results. SD has no annotation that explicitly states that cases are only confirmed, but we started reporting values in positive cases (PCR) on 7/17. These values should be removed from positive cases (PCR).

Source: https://doh.sd.gov/news/Coronavirus.aspx

State or US: South Dakota

Describe the problem Unclearly defined case count included in confirmed cases. SD has unclear case definitions.

Link to data source https://doh.sd.gov/news/Coronavirus.aspx

Death values are historically recorded in both the "Deaths" and "Deaths (Confirmed)" columns for SW. However, SW’s death values are unclear about what they represent, so they should only be recorded in the main "Deaths" field.

Visiting https://covidtracking.com/api/graphql in chrome or using the API Playground (https://covidtracking.com/api/playground) produces the following error message rather than data.

Error: Interface field DataPoint.positives expected but USDailyDataPoint does not provide it.

Interface field DataPoint.negatives expected but USDailyDataPoint does not provide it.

Interface field DataPoint.positivesPlusNegatives expected but USDailyDataPoint does not provide it.

Interface field DataPoint.deaths expected but USDailyDataPoint does not provide it.
    at m (worker.js:29:129250)
    at x (worker.js:29:6581)
    at E (worker.js:29:5626)
    at Module.O (worker.js:29:5365)
    at Object.t.generateSchemaHash (worker.js:51:56541)
    at l.generateSchemaDerivedData (worker.js:51:147)
    at new t.ApolloServerBase (worker.js:39:274246)
    at new l (worker.js:73:19678)
    at worker.js:39:128673
    at a (worker.js:39:128736)