Many years ago I started out as a Chem major. After 3 years and many hours I quit.
My grades were swell, I still loved science - but I couldn't be honest with myself that I was going to be a real scientists in the end.
So here's what typically goes wrong: You get a lab assignment. Somewhere along 4th hour or second week you screw up and grab the lab ass for help. He says "sorry - just keep going."
You won't get a do-over. You can't afford to start over because the expected competence requires immediate good results.
Your lab time is limited and resources are scarce. If you want the grade you'd better "learn from your mistakes" and "be more careful next time."
What about the results? Well you already have a pre-conceived notion of what they should be. Maybe you get them from your mates, or look them up.
Learning from our mistakes is "science code language" for pushing small known nudges and data "massaging" as an acceptable method for passing the course.
Perhaps at one time students really did learn and grow from these mistakes - but the modern concept of failing is simply a quick exit from a highly competitive major.
All the while my own scientific rigor which I was supposed to be enforcing on my own results was slowly corrupted.
I couldn't truly say that my practise in the scientific method was honestly the truthful result of my own observations and methods. Perhaps I was just too anal at the time - so be it.
I worked in an environmental testing lab and the sloppy procedures practiced were sometimes much the same.
Circumstantial? Yes. Take of it what you will. But if my experience is like that of others - such corruption as reported here doesn't surprise me.
Science should be about failure as much as it is about success. Failure is a valid result - but we often fail to oblige real and honest failure as scientifically and (most importantly) educationally valid.
That was my draw to CompSci. Our very embrace of failure as a tool for learning. I absolutely love it.