People are applying a stronger point here where changing the original data is hard evidence of foul play and sufficient to completely discount any changes.
We'd be grappling with the reality of faster than light neutrinos now, though, if that was correct logic.
Manipulating field measurements is already frowned upon in all applications. There is no point. Things like data provenance is a serious issue, which is directly targeted by peer reviews and investigations on scientific malpractice. Even performance benchmarks highly favor standardized test and data sets.
Being objective matters. Once you start messing with original measurements, you place yourself in a position where you need to answer questions on whether you're just adapting data to fit your belief instead of the other way around.
How do you know this? Do you understand the reasoning behind bias correction?
> Do you understand why this is a problem?
The problem is people do not understand bias correction and you are certainly not arguing on a methological level.
It's not reductionism, and you're missing the point.
The whole point is that changing the original data to be able to support/reject hypothesis naturally raises the question of whether there is foul play.
If original measurements don't support your claim and suddenly by changing data you get it to fit your belief, any rational analysis would quickly flag the risk of data manipulation and scientific malpractice.
If instead you're dealing with a politically charged topic that attracts denialists and contrarians then you're making yourself vulnerable to accusations of fraud that will certainly be used to poison the well.
Do you understand why this is a problem?