Chemistry has its fair share of examples, from plagiarism to crystallography fraud. You have to realise that often research is very specialised and it will often take significant time and resources to prove particular data false. It is very easy for that researcher to say 'you are doing it wrong' than admit issues with the data.
But my major point wasn't so much about outright fraud, as little lies - cherry picking of data to make it sound more significant is rampant.
I get that chemistry is specialized - all fields of science pretty much are now-a-days... but the specialization doesn't lead to cherry picking of data... For example it is near impossible to 'cherry pick' your data if you're in a synthetic lab.. when someone goes to repeat what you've claimed in your paper they will see that you are full of it and call bullshit... if you can't reproduce the compound you claimed you made and show them how with similar yields your stuff will be getting retracted... the same is true for my computational work... I can't just fake my data - when someone else runs my simulation with my specified parameters they will see that the values I claim I'm getting are completely bogus... Of course since my work is also highly theoretical there are areas where we just can't be sure if the physics are right but that of course is always appropriately accounted for whenever we publish our results or whenever applying for a grant etc...
So yes there is room for dishonest people in science to mislead because the very nature of scientific discovery requires that we speculate...
but to say that science is rampant with fraud just because you can find a few examples - is like saying that all men rape women just because you can find a few examples of men that do - thereby turning rape in to a problem with males as opposed to a problem with individuals who commit a crime..
As I said in my earlier post - it's not that science itself is flawed but rather that people are flawed - and as I said before - of all the fields - and these people enter all of them - science is better equipped to handle dishonest people than any other because unlike business or law or whatever else our work is independently reviewed and eventually you will get caught...
you talk about science like other researchers in your area will have no idea if what you're saying is likely bullshit or not and it just doesn't work that way..
Research labs don't exist in isolation working on stuff that the rest of us cant possible understand.. I can easily look at another lab's data and call bullshit as can most other scientists... and in the cases where you can't - well - some things are so theoretical it's hardly fair to call incorrect assumptions purposefully misleading when in fact they were simply wrong....
Science is highly specialized sure - but not to the point that you're suggesting... where one person or one lab are the only ones who can comprehend what they're working on - this isn't the 1700s or whatever.... It is not as easy to fake data, cherry pick data, etc as you claim here...
Especially in the natural sciences... because crystallography requires human input it is easier to 'fudge' data in that area sure, and if you're especially dishonest you can attempt to mislead the lazy grant/journal reviewer - but you will eventually get caught (as the people you're referencing here apparently did)... it isn't as if each crystallographer is off in his own lab fudging assignments and manipulating algorithms to produce geometries that don't actually exist in his/her compound or that even if they were you wouldn't be able to easily ascertain this via carefully looking over their original data or repeating their work on your own.. Remember that one can always request the original crystal data or simply grow the crystals in question and compile their own data to see if it matches...
So yeah - some people fudge/cherry pick data but they hardly comprise the majority... and there are some areas of science - especially in my line of work - where fudging the data isn't really possible... if I say I made a compound I have to turn in the NMR proton, carbon, and any other relevant nuclei, (1D&2DNMR data + optical rotation) if it is chiral and they always are, elemental analysis, mass spec, GC <--- and that is just for the compounds that I made to prove the computational data I obtained from molecular modeling of whatever transition state I'm claiming results from whatever mechanism I'm claiming my reaction goes through to produce the thing I claimed it would that I then have to prove that I have via the methods I already outlined above..
I don't think it's even close to accurate to claim that 'cherry picking' of data is rampant throughout science just because a few people have done so... for every lab you can find that is doing this I can find about 20 more who are not... Dishonest people exist everywhere and fudging data or cherry picking data is a problem with the individual not a problem with science.. again - going back to the rape example - blaming money/fame whoring in science for people fudging data is like blaming scantily clad women for getting raped... it's not the clothing or the money it's the people and they are everywhere in all fields and when the people who are put in place to ensure the rules of the system are followed get lazy or for whatever reason have limited resources and time dishonest people will always find a way to slip through the cracks - but what typically happens when they try this in science - is they get caught because our work builds on previous work and if the foundation is not strong eventually it all crumbles and it typically isn't hard to figure out where the cracks were when you start looking etc... and you will either find someone who was innocently mistaken or someone who just flat out made their shit up...
*shrugs*
but that's life