4 ms·
I'm cynical and have no trouble believing that results were twisted immensely prior to the rule change in 2000. The incentives are huge: profit, prestige, or si
by caymanjim 8y ago
I'm cynical and have no trouble believing that results were twisted immensely prior to the rule change in 2000. The incentives are huge: profit, prestige, or simply job security.
I do wonder if this chart misrepresents something, though: there are studies that produce incidental--but genuinely valuable--discoveries. It's unclear to me if that accounts for the pre-2000 results or not. With the new rules, would there have to be another study stating the new objective?
I'm not questioning that bad science occurs, but I am questioning what this graph really tells us.
- rdiddly 8y agoYou're right, the new rule doesn't filter out dishonest discoveries, only incidental ones (which may be dishonest). If you wanted to show benefit from an incidental discovery, you would have to run another study with that as its declared objective. The author is kind of doing the same thing here that they accuse others of doing...
- scotty79 8y ago> With the new rules, would there have to be another study stating the new objective? I think it's only fair to force you to replicate at least once the positive result you think you see in the data you collected for another purpose before you can claim you got something.
- Balgair 8y agoYes, the linked article is saying that researchers used to employ techniques like 'p-hacking' (among others) in order to report results that were favorable/novel. That the scientists and clinicians in charge teased the data too much.