As a laywoman myself - Does not having wide access to studies that show null results lead to lots of wasted effort? Like, a dozen research teams all trying the same thing in vain without knowing...
As a laywoman myself - Does not having wide access to studies that show null results lead to lots of wasted effort? Like, a dozen research teams all trying the same thing in vain without knowing anybody else has already tried and gotten null results? Seems like that's as useful to know as whether something produces positive results.
That's one problem, but it's hard to know its extent, precisely because most of these efforts are never published. Another (bigger imho) one is that p-value hunting decreases the confidence you...
That's one problem, but it's hard to know its extent, precisely because most of these efforts are never published. Another (bigger imho) one is that p-value hunting decreases the confidence you can have in studies that show positive results. Pre-registered studies solves both problems to a large extent. Also, peer-review before running experiments sounds like a very good idea.
It's refreshing to see an article written by people familiar with science. Not only does it lay out the actual findings, rather than sensationalize them, but it puts it in context and explains...
It's refreshing to see an article written by people familiar with science. Not only does it lay out the actual findings, rather than sensationalize them, but it puts it in context and explains some additional possibilities for interpretation.
In regards to pre-registered trials, having an analysis like this that shows they have drastically different results is a huge step in the right direction and hopefully will help speed up adoption.
In some fields, such as exercise and nutrition sciences, I hope the adoption is even faster because they are in grave need of reform and heavily polluted by garbage science.
As a laywoman myself - Does not having wide access to studies that show null results lead to lots of wasted effort? Like, a dozen research teams all trying the same thing in vain without knowing anybody else has already tried and gotten null results? Seems like that's as useful to know as whether something produces positive results.
That's one problem, but it's hard to know its extent, precisely because most of these efforts are never published. Another (bigger imho) one is that p-value hunting decreases the confidence you can have in studies that show positive results. Pre-registered studies solves both problems to a large extent. Also, peer-review before running experiments sounds like a very good idea.
I'd never heard of p-value hunting before today but I looked it up. Yeah, this definitely sounds like a move in the right direction.
It's refreshing to see an article written by people familiar with science. Not only does it lay out the actual findings, rather than sensationalize them, but it puts it in context and explains some additional possibilities for interpretation.
In regards to pre-registered trials, having an analysis like this that shows they have drastically different results is a huge step in the right direction and hopefully will help speed up adoption.
In some fields, such as exercise and nutrition sciences, I hope the adoption is even faster because they are in grave need of reform and heavily polluted by garbage science.