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

North Dakota negative deaths 11/02/2022 and Nebraska negative cases 10/26/2022 #6250

Open CSSEGISandData opened 2 years ago

CSSEGISandData commented 2 years ago

Hello all,

We are aware of the following anomalies in our dataset. Please see below for explanations.

For both Nebraska and North Dakota (deaths), our sourcing is via the HHS Community Profile Report (CPR) for county-level cases and deaths and the CDC COVID-19 Data Tracker for state-level cases and deaths. Any differences between the HHS CPR and CDC Data Tracker are placed into our "Unassigned, STATE" category. We use these sources for the two states due to Nebraska discontinuing their dashboard and North Dakota discontinuing their reporting of deaths.

Decrease of -155 deaths in North Dakota on November 2, 2022 We have tracked this to a decrease in the HHS CPR published on November 3, 2022 (Community_Profile_Report_20221103_Public.xlsx). Historical versions of the CPR are available at the above link by clicking "Show more" at the bottom of the page. The decrease is a nullification of the deaths published in the "Unallocated, North Dakota" entry, which were previously 155. The reason these are dropped is unclear. As the historical files contain these deaths, we will maintain our historical reporting and continue to reflect that which is published by the HHS CPR.

Decrease of -8099 cases in Nebraska on October 26, 2022 We have tracked this to a reorganization of the data published on the CDC COVID-19 Data Tracker. On this day, cumulative probable cases in the state (164193) were folded into the confirmed cases entry for the state (376152) to create a new total for confirmed cases (532246) that was less than the total of the two columns before they were combined (164193+376152 = 540345). The reason for this difference is not clear but has persisted. Prior to this change, the CDC COVID-19 Data Tracker consistently reported more cases than the HHS CPR which was the reason for the accumulation of cases in the "Unassigned, Nebraska" entry. With this change, the number of cases in that entry fell to zero and has been steadily climbing since.

Hansonyao417 commented 2 years ago

Hi, I am currently a Senior at University of Michigan Majoring in computer science. I am currently taking a course call software engineering, the final assignment is to solve an open source problem. I would like to work on this issue for the next few weeks and attempt to solve it. My number is 2318831169. Thanks!

anthonydb commented 2 years ago

@CSSEGISandData Do you have any indication why there haven't been any North Dakota deaths reported since that -155 decrease you've noted above?