LogFAQs > #910557905

COVxy posted...

scar the 1 posted...

When a paper seems to have an experiment with actually interesting results, only that the conclusion is way out there, how do you judge that? Because that's the case here, they mix plausible data with outlandish reasoning.

I would reject it in the position of an editor. If the conclusions don't match the data, that's a severe flaw.

If I ran a pure behavioral study on visual detection, and then concluded about thalamocortical basis of attentional tuning, it would get desk rejected. Almost certainly. You can't just conclude anything if you have "data". Scientific reasoning is making reasonable conclusions in the light of data. That's all science is.

Yeah so where do you draw the line? If some of the conclusions are wild and some aren't? Etc. Again, the goal of these papers was always to make them seem as publishable as possible, not to make them look clearly and obviously fake. Most of them were rejected, and several of them got severe criticism in the review stage.

But yes, of course an editor should clearly reject someone who has a conclusion that is completely disjointed from the results.
---

Everything has an end, except for the sausage. It has two.