How Do Scientists Faking Their Data Affect All Scientific Research

Scientists faking their data affects all scientific research. Most experiments are based off of other experiments that have already been done and if there's false data then it could change the conclusion of the tests and it could alter the results. If scientists didn't "fudge" their data then there could be more cures for diseases. Technology could advance and new discoveries could be made. The counterargument to that statement, though, is that if scientists only change their data a little bit on the small stuff, they can get bigger grants to make more important discoveries. But how does a person measure the "small stuff" and the "big stuff"? I think that it doesn't matter when a person lies or fudges data, because out of their small lie could come another experiment that could turn out to be false because they based it on someone else's data.
If a student in a classroom sees that the person next to them got a different answer and they change it so it matches the other person's answer, they're not learning anything. Data varies and the fact that the student changed it to try and get a better grade is really sad. Also, if the student changes their results so they match their hypothesis then it's not helping them learn anything. When scientists change their results to match their hypothesis it's not doing them any favors either. If anything they should change their hypothesis instead of their data. If a scientist looks at the data and the first thing they think of is "what do I want the data to tell me" they're looking at all of science entirely wrong. It's not about what you want the data to tell you, it's about what it does tell you, and what it means. If scientists selected just the points that prove their hypothesis they're being completely unethical and misinforming the public and other