Twitchy favorite AG blasted The Hill, and other outlets for continuing to mislead readers on the “results of increased testing” which pushes the number of cases up in a state but at the same time ignores that the percentage of people testing positive has gone down, which is a good thing:
This is misleading on several levels.
1) New cases are the result of increased testing. Positive rate actually went down over the same time. Punishing TX 4 testing more.
2) Clearly unrelated to re-opening as any resulting spike wouldn't be identified for at least 1-2 weeks. https://t.co/cUz8dyjMZU
— (((AG))) (@AGHamilton29) May 7, 2020
It’s also irresponsible to tie this increase to states that just started to open up this week:
Outlets keep doing this, and it's a very bad look. You can't demand more testing and then attack states with misleading data that results from them doing it. Further, it's obviously impossible for new reported cases today to be result of policy actions a day or two ago.
— (((AG))) (@AGHamilton29) May 7, 2020
But don’t take AG’s word for it, even though you should. Here’s Nate Silver making the same exact point:
One basic mistake is in not accounting for tests. If a state doubles its testing it is going to see far more cases. Many articles about big spikes don't even bother noting when they were caused by tests increasing. The math isn't hard. It's a basic failure to provide context.
— Nate Silver (@NateSilver538) May 5, 2020
Recommended
The media is “actively misinforming people and creating discentives to do more tests”:
If you greatly expand testing—something that is CRUCIAL to reopening SAFELY—and then a bunch of news articles slam you because your state is finding more cases as a result of doing more tests, they're actively misinforming people and creating disincentives to do more tests.
— Nate Silver (@NateSilver538) May 5, 2020
And notice how the media is still singling out red states? Here’s the Mayor of Denver explaining how the “uptick in coronavirus cases correlates with increased testing in the city”:
Denver's @MayorHancock tells @PoppyHarlowCNN the uptick in coronavirus cases correlates with increased testing in the city.
“We look at the full spectrum of the impact of Covid in our communities before we make decisions to open them up,” he says. https://t.co/YUZ5GbgT1i pic.twitter.com/9HXYnmeeOv
— CNN Newsroom (@CNNnewsroom) May 7, 2020
Want more misinformation? Check out this chart via MSNBC’s Jesse Rodriguez. This charting just the increase in positive tests AND not using per capita numbers:
U.S. curve isn't bending @JohnsHopkins @axios pic.twitter.com/8dAtuH7tRm
— Jesse Rodriguez (@JesseRodriguez) May 7, 2020
We need the “RATE”:
With increased testing, what's the actual RATE of positives though? That's a more useful metric.https://t.co/raXfroGauS
— Dan Goldwasser (@dgoldwas) May 7, 2020
“Re-graph” that and tell us what you get:
Please make it more selective, take out NY, NJ, MA, PA and CA, then re-graph
— GaryB (@yragnia) May 7, 2020
Bracket out each state and see what it looks like. NYC-based journos won’t be happy, however:
Really have a hard time with this thinking. Either the US is a thing epidemiologically or it’s not. If you can bracket out NY, shouldn’t you also bracket out (say) CA or other regions? That curve is a composite of other curves that are probably radically different from each other https://t.co/FxhBRTCAsi
— David Hazony (@davidhazony) May 7, 2020
***