Every Monday, once the Seattle Public Schools COVID-19 data dashboard has been updated, we publish an update listing the changes in local schools’ cumulative case counts. This past Monday, one school – West Seattle HS – showed a startling change, with the new cumulative count more than double what was shown a week earlier. Since we hadn’t seen an anomaly like that before, we asked the district about it. Here’s the explanation from SPS spokesperson Tim Robinson:
The issue here is one of a reflection of data availability, not trend shifts.
The dashboard was updated 1/24. The next day, dozens of rows were added to the WSHS database used by nurses entering the information. Consequently, the dashboard reflected older cases that had not yet been entered, which gave the appearance of a significant case jump.
The actual data from WSHS tells a positive story of decreasing cases. WSHS week-to-week data:
9/4/2021 – 9/10/2021: 1
9/11/2021 – 9/17/2021: 1
9/18/2021 – 9/24/2021: 1
10/23/2021 – 10/29/2021: 1
11/20/2021 – 11/26/2021: 1
11/27/2021 – 12/3/2021: 1
12/11/2021 – 12/17/2021: 5
12/18/2021 – 12/24/2021: 10
12/25/2021 – 12/31/2021: 3
1/1/2022 – 1/7/2022: 15
1/8/2022 – 1/14/2022: 55
1/15/2022 – 1/21/2022: 39
1/22/2022 – 1/28/2022: 7
Dashboard users can see and adjust the week-to-week data by using the sliding date filter in the upper left corner of the dashboard. (Pro tip: click on the date rather than use the slider, as that brings up a calendar that allows you to enter the specific date.)
Lastly – please see the note in red type at the bottom of the dashboard: “Due to high community transmission, contact tracing efforts have been prioritized to focus on notifying confirmed cases. Data continues to be processed as it becomes available so counts displayed here are subject to change.” While there may be a data lag on the dashboard, necessary action of communicating to those who require information is prioritized and does not lag.
The end dates on the time spans above are Fridays – each Monday’s dashboard update reflects case counts through the preceding Friday. Comparing the totals shown in the two previous weeks to the adjusted data, the January 24th increase was 30 below what it should have been, and the January 17th increase was 50 lower than the real number.