Questionable Claims about Northern California Risk from The Guardian
After largely avoiding the worst of the pandemic, a block of far northern California counties now leads the state with nearly 40 cases per 100,000 residents over the past week, according to statistics maintained by the Los Angeles Times. Tehama county ranked the highest in the LA Times case ratings with 139 cases per 100,000 residents. Meanwhile 10 of the 21 total Covid deaths in nearby Siskiyou county have occurred since the beginning of May.
If true that's pretty alarming, so I decided to investigate. Doing so made me wonder if The Guardian understands the numbers it is reporting.
As noted previously on this blog, there are wrong numbers and right numbers to use when you're trying to make comparisons between geographic areas. It is important to use population adjusted numbers (usually cases per 100K) and it it also important to use moving averages (usually over the past seven days) to smooth-out irregularities in reporting.
I looked at the LA Times' dashboard, and I can't for the life of me figure out where The Guardian got the numbers they quoted. At the moment, their main new cases map is not reporting any data at all from Tehama County. A link for the county says "Over the last seven days, officials have reported 44 new cases, which amounts to 69 per 100,000 residents."
So first off, that's half of the 139 reported by The Guardian. That is suspiciously close to twice the value of 69 per 100K. It looks to me like they were taking their data from the main map which gives figures for two weeks—but, again, it is currently reporting no data for Tehama.
If you take the last quote literally, it is also a cumulation over 7 days, not an average. If we divide the by seven, we get 9.86 per 100K, 7-day average. Is that terrible? As reported in this blog on Sunday, Maricopa county in Arizona has a rate of 9.66, so it's about the same as here.
The LA Times site says Los Angeles County reported an average of 720 new cases over the last week. Adjusted for population that is 6.89 per 100K (though the adjustment should be done prior to the average so that might be somewhat off the true per 100K 7-day average).
It certainly would be better if Tehama County were not 50% higher than LA County, but three cases per 100K doesn't seem like anything to panic about. It's certainly not as bad as the article makes it sound. The Guardian needs to improve it's data journalism practices.
Image by David Mark from Pixabay