raginrayguns:

a lot of scientists don’t act on scientific ideals; a lot of them are actually bad people, who in both their professional and private lives are trying to get whatever they can get away with.

This is something I haven’t mentioned even when writing or talking about something closely related, which is why you shouldn’t trust the conclusions of a scientific study. I mean, a single scientific study. Because, I say, a single scientific study is typically using a single model, or studying a single population; you don’t know if the results really generalize. Or there could have just been some honest mistake; my go-to example is the paper about the faster-than-light neutrino. One of my friends, when I broke the news (apparently he hadn’t heard) that this turned out to be an artifact, he was kind of angry at the authors. But, although I don’t know the details, I don’t think they did anything wrong; my understanding is they made the measurement and they reported it. What would have been wrong would be to not report the measurement because they know the conclusion is impossible. So, even when everyone is acting as they should, sometimes the conclusion of a paper is misleading. That’s why we have reviews and meta-analyses, which can cite hundreds of studies. That’s how we know what’s really a general principle, that’s how we can make statements about what’s happening in the real world instead of one laboratory or one population.

But it must also be remembered that everyone is not acting as they should, so you have even more false and misleading results than you would in the best case scenario. These people will happily publish results that do not meet the standards of evidence of their field, if no one is forcing them to meet the standards of evidence of their field, and reviewers usually do not catch everything.