obviously you cannot include social distancing as a control as it is potentially affected by treatment, luckily the paper does more. Now please stop spamming unfounded nonsense confirming your priors and look at the actual paper
RCT proves masks work
-
ok undrgreeads. Think of an RCT for some medication. A placebo would be a group that gets a pill with nothing in it, whereas a control group could simply be a group that gets nothing at all.
Yes, exactly. So what would a placebo mask treatment look like - something that looked and felt like a mask, but was actually full of holes? Ridiculous.
People who demand a “placebo group” for masks like that ElonBachmann re.t-.rd above simply reveal they were going to be anti-mask no matter what.
And every time a Chinese person sees an anti-mask tweet, they say “these Westerners are crazy!” and the CCP gains another year in power.
-
https://www.washingtonpost.com/world/2021/09/01/masks-study-covid-bangladesh/?utm_source=reddit.com
File this under “things everybody already knew”Treatment village sero = 0.69%
Control = 0.76%
Tiny effect with a p-value just sneaking in under 5 per cent which I am absoultely sure is not a function of data fiddling.They have a pre-registered experiment which considerably reduces opportunities for data fiddling.
-
Even with a p value of .1 the implication of this study is mask wearing works, even with a marginal improvement in compliance. Surgical masks work better than clothe makes. But they both are better than nothing. It might be hard to Disentangle the effect of social distancing, but again, the suggestion is the current guidance works. Wear a mask. Keep your space.
Weird ethics of RCT’s aside, this is an impressive piece of research.
-
Even with a p value of .1 the implication of this study is mask wearing works, even with a marginal improvement in compliance. Surgical masks work better than clothe makes. But they both are better than nothing. It might be hard to Disentangle the effect of social distancing, but again, the suggestion is the current guidance works. Wear a mask. Keep your space.
Weird ethics of RCT’s aside, this is an impressive piece of research.yes, JA is making some serious waves with this one
-
This. Undergrads often confuse being cynical for being insightful.
Not saying the effect sizes are overwhelming with this study but at least inform yourself before pulling out the "I'm so sophisticated in my thinking that I can see through the fraud perpetrated by grownups" card.Treatment village sero = 0.69%
Control = 0.76%
Tiny effect with a p-value just sneaking in under 5 per cent which I am absoultely sure is not a function of data fiddling.They have a pre-registered experiment which considerably reduces opportunities for data fiddling.
-
This. Undergrads often confuse being cynical for being insightful.
Not saying the effect sizes are overwhelming with this study but at least inform yourself before pulling out the "I'm so sophisticated in my thinking that I can see through the fraud perpetrated by grownups" card.Treatment village sero = 0.69%
Control = 0.76%
Tiny effect with a p-value just sneaking in under 5 per cent which I am absoultely sure is not a function of data fiddling.They have a pre-registered experiment which considerably reduces opportunities for data fiddling.
hear hear
-
Kind of hilarious that Uhlig of all people asked for an RCT as evidence that masks work. Sure, great we got an RCT now. But it was always very plausibly that masks, especially good ones, work. The guy probably cannot even spell credibility. Sad to see him turn into a right wing hack
-
Kind of hilarious that Uhlig of all people asked for an RCT as evidence that masks work. Sure, great we got an RCT now. But it was always very plausibly that masks, especially good ones, work. The guy probably cannot even spell credibility. Sad to see him turn into a right wing hack
‘Masks work because a few economists tortured a dataset until they could show statistically significant, though not economically significant, differences between ONLY certain treated and control groups.’
-
Kind of hilarious that Uhlig of all people asked for an RCT as evidence that masks work. Sure, great we got an RCT now.
The CDC put out a meta analysis of 10 surgical mask RCTs in May 2020 that showed they did not significantly reduced the transmission of influenza.
-
Even with a p value of .1 the implication of this study is mask wearing works, even with a marginal improvement in compliance. Surgical masks work better than clothe makes. But they both are better than nothing. It might be hard to Disentangle the effect of social distancing, but again, the suggestion is the current guidance works. Wear a mask. Keep your space.
Weird ethics of RCT’s aside, this is an impressive piece of research.In the context of the country the study is conducted in, this study is a bit like moving people away from bad american diet, putting them on a paleo, and showing they lost weight.
Moreover, these numbers do not mean anything for US policy.
-
On this website, there are lots of attacks on "regmonkeys" - which I take to mean people who run simple regressions without any deep understanding of economics. Some of the comments on this thread suggest that there are economists (if indeed that's what the commenters are) who don't even understand simple regression, or even simpler RCTs that don't even require running regressions. See, for example, the comment about "control for social distancing". There are also economists who appear to lack common sense - e.g. comments about "where's the placebo".
I propose that you all find some other phrase beyond "regmonkey" to refer to the truly idiotic economists amongst you.