3 ms·

I don't know much about the study quoted but I wanted to make an general counter-argument to one of your statements. I often see this counter-argument made impl

I don't know much about the study quoted but I wanted to make an general counter-argument to one of your statements. I often see this counter-argument made implicitly but for the sake of clarity I'd like to explicitly state it here. >And if this particular study indeed turns out to be bad (not replicating, bad methodology, etc) that would validate the above truth even more. Publishing bad-science and then saying other scientists will show it is wrong is both true and does not address the central complaint: it will cause harm to those outside the scientific community. This is because as far as the public is concerned publication functions as endorsement. For instance the Antivaxxer movement got an enormous boost when incorrect science was published that purported to show that vaccines cause autism. It is likely that publishing that study has resulted in many human deaths. If a paper makes a claim which could cause public harm, that study should be held to a significantly higher standard of evidence than a study which is unlikely to cause harm. Consider two papers: 1. A paper which argues that photons travel faster than the speed of light. 2. A paper which argues that diet coke cures cancer. They both might be sloppy research, they both make extraordinary claims but the second ones harm to society is higher and thus must have very strong evidence to be published.

coldtea 7y ago
>Publishing bad-science and then saying other scientists will show it is wrong is both true and does not address the central complaint: it will cause harm to those outside the scientific community. If "publishing bad science" was the issue people were concerned with, then 80% of published (peer-reviewed and all) papers should be met with the same or worse reactions for they are time and again shown to be bad science. But I fear that it's the "we don't like the conclusions" more than the "bad science" part. And I fear that the latter (which might very well be the case) is even a priori assumed given the conclusions. In you response you take for granted what oughta be proved: that the paper would cause "public harm".
- Vinnl 7y ago
  GP did not argue that "publishing bad science" was a concern, but the effects of the publication of bad science. Those are to be considered greater and worse here than with 80% of published research.
- EthanHeilman 7y ago
  >In you response you take for granted what oughta be proved: that the paper would cause "public harm". Yes, this is exactly my point: If for any paper we conclude that it would cause "public harm" then the burden of evidence should be increased. The above post does not argue that paper does or does not cause harm but that all criticism of publication is void because later papers would show the original to be incorrect. This is a very strong claim.

coldtea 7y ago

>Publishing bad-science and then saying other scientists will show it is wrong is both true and does not address the central complaint: it will cause harm to those outside the scientific community. If "publishing bad science" was the issue people were concerned with, then 80% of published (peer-reviewed and all) papers should be met with the same or worse reactions for they are time and again shown to be bad science. But I fear that it's the "we don't like the conclusions" more than the "bad science" part. And I fear that the latter (which might very well be the case) is even a priori assumed given the conclusions. In you response you take for granted what oughta be proved: that the paper would cause "public harm".

Vinnl 7y ago
GP did not argue that "publishing bad science" was a concern, but the effects of the publication of bad science. Those are to be considered greater and worse here than with 80% of published research.
EthanHeilman 7y ago
>In you response you take for granted what oughta be proved: that the paper would cause "public harm". Yes, this is exactly my point: If for any paper we conclude that it would cause "public harm" then the burden of evidence should be increased. The above post does not argue that paper does or does not cause harm but that all criticism of publication is void because later papers would show the original to be incorrect. This is a very strong claim.