Hacker News Reader

Comment by epups | Hacker News Reader

As someone with an academic background, I would say it is very easy to forge results without anyone knowing it, at least in the biomedical field. The cases being caught are usually lazy like literally copying and pasting bands, if even a medíocre attempt to cheat was made it would have been undetectable. In many cases the professors involved in outright fraud don't even lose their career.

View on HN

Crazy that we only have Elizabeth Bik as the one person who tackles this kind of fraud. You’d think NIH would care more about where their billions of dollars go every year

I grinned when I saw "biomedical". For some reason my mind always connects to grand old Gregor Mendel, and the Mendel Fisher controversy. I thought you might be subtly trolling, in a really clever way. I believe your comment was made in earnest.

Trusting results it tough, and only tougher with scale. Regardless, brilliant comment even if the link was unintentional.

Honestly I wasn't thinking about Mendel here. That's a big exception of course. I feel that at least some people who cheat imagine themselves to be doing something similar, cleaning up data to show a deeper truth. In reality, they cost science and individual researchers a lot of time and money.

Of course it's a dire problem which is not nearly taken serious enough. Interestingly enough, some fields seem to have some intrinsic motivation to replicate, e.g. biology [1]. I hope we can create similar incentives for all sorts of science.

(caveat emptor: I don't know anything about biology)

[1] https://statmodeling.stat.columbia.edu/2022/03/04/biology-as...

Also, forging results is just one variety of unethical practice that is incentivized in academics today, out of many possibilities. There are many other things to do that involve greater degrees of plausible deniability.

Reply on news.ycombinator.com