This is just way too true. I know someone (anecdote alert) learning at a really prestigious university in a medical field and she told me multiple times that they intentionally cheat the results to match the expected output.
By cheating I mean... flat out lying. I don't know the implications of this (how far misinformation can get) but it seems like a wrong culture and attitude, especially for science.
Such a thing happened in my graduate school lab. I'll even cop to having a data point in one graph where I just got sick and tired of doing the experiment so there's an N of 4 instead of an N of 5 as it states it is in the methods section (this is actually impressive for the field I was in at the time, which typically didn't even do replicates at all. I am sure my colleagues' results are, at best, cherry-picked).
You know, I did this in my freshman chemistry labs. I mean, sometimes I just made up the numbers for my "observations", varying them by some amount from the predicted numbers.
At the time I definitely wondered how much of that went on in "real" science. I suspected (and still suspect) it's a lot more than is ideal.
By cheating I mean... flat out lying. I don't know the implications of this (how far misinformation can get) but it seems like a wrong culture and attitude, especially for science.