Here's a lot of people's experience with science labs: http://pages.cs.wisc.edu/~kovar/hall.html http://pages.cs.wisc.edu/~kovar/hall.html It's not uncommon fo

Here's a lot of people's experience with science labs: http://pages.cs.wisc.edu/~kovar/hall.html http://pages.cs.wisc.edu/~kovar/hall.html It's not uncommon for students to admit to "fudging the data" so they get the mathematically-aligned results of the theories they were taught first say should happen, so that they can write their results paper showing that indeed the experiment hints at a simple math law, and go home. And some professors will intentionally bias equipment to produce faulty data in order to fish out the fudgers. But mostly they don't need to go that far as the first link shows, just don't supply reliable equipment, or neglect to point out that even with very reliable equipment it still isn't enough to get reliable data, or even the right kind of data in principle (the material setup you selected might not allow for finding the thing you want to find [1]). The "history" aspect of it is perverse at times because you're learning a series of slightly less wrong lies at each step. Lots of chemistry classes go through every model post-alchemy and superficially examine some of the experiments that seemed to accord with or violate the existing prevalent model, but that doesn't help teach scientific thinking, and is all rather a useless set of information most people will forget anyway (just like the history courses full of wars and their dates). It's really not a good idea to teach science as a series of lessening lies, but that's the way it's commonly done. Lots of undergrad intros to E&M will barely touch on Maxwell's equations at the end. Whereas Feynman's book at least has the decency to talk about the reality of those equations up-front, then explicitly talk about the more-wrong simplifications while keeping the reality at the back of the mind until a bit of domain intuition is built up. In quantum physics apparently (http://oyhus.no/QuantumMechanicsForProgrammers.html http://oyhus.no/QuantumMechanicsForProgrammers.html) you get things like trying to mathematically derive "the Dirac wave function from the Schrodinger wave function combined with spin and relativity" and an idea that you can just math your way through things. [1] https://dreamsongs.com/Files/Incommensurability.pdf https://dreamsongs.com/Files/Incommensurability.pdf On page 17 relates the story of how two different material setups in experimental design led to contradicting evidence for the existence/non-existence of free quarks. The footnote pointing to Feynman's comment about an earlier experiment (https://en.wikipedia.org/wiki/Oil_drop_experiment#Millikan's_experiment_as_an_example_of_psychological_effects_in_scientific_methodology https://en.wikipedia.org/wiki/Oil_drop_experiment#Millikan's... and the whole "cargo cult science" text of his in general) is worth a read too. Then there are entire pseudo-scientific fields like parapsychology (https://www.lesswrong.com/posts/enuGsZoFLR4KyEx3n/parapsychology-the-control-group-for-science https://www.lesswrong.com/posts/enuGsZoFLR4KyEx3n/parapsycho...) that use all the same techniques actual scientific fields like psychology use. In light of the replication crisis it's all the more interesting to note because knowledge in say probability theory or the scientific idea of creating (non-statistical) models of a system that could have lessened the crisis is just not taught. This sort of stuff was never covered in my science education at any level. I think some (all?) might be covered in "philosophy of science" types of coursework but I never took any, and have seen some indication that it might be a good thing as perhaps many of them devolve into promoting an antirealism point of view. Anyway this all ties into the GP's complaint that scientific thinking and process just isn't taught very well. Given the population's advances in STEM in spite of that, though, perhaps it's just not very important to complain about.