Interesting update on this paper published that purported to show that staying at home doesn't reduce COVID-19 deaths: less than a week after publication it already has a warning from the editorspic.twitter.com/NZbn54Oja3
You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
Love it. We feel our data are slightly above publication quality. Many will not agree w/this as it is close to the border.
I'm just baffled. It's trivial for them to rerun the code on another dataset, but I should download all of their work and spend hours recreating it instead?
The other major issue with this study is there is no apparent attempt to account for lags between "staying at home" and deaths, nor to separate different directions of causality. It seems the study is looking for an association between how much staying at home happens in country
A in week t and deaths in country A *in the same week*. Even if the google data were perfect, any effect of staying at home on deaths would be lagged by several weeks. On top of that an increase in deaths could drive an increase in staying at home which will confound any attempt
Can they redo their analysis to look at the reproduction number vs mobility? Even for the not-great home category there's a clear correlation (black line is adding those together; using negative of home category)... seems to obvious to write a paper but apparently not!pic.twitter.com/2iyfR8pNQ6
And if their data cannot address the ‘question’ as they erected it, WHY have such a blaring paper title that says the opposite? I blame the journal don’t know what they were thinking. When a ‘result’ appears so contrary they should’ve been more careful.
To me, the response is the academic equivalent of “I’m rubber, you’re glue”, no?
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.