Study shows, "No racial bias in police shootings".

[anigbrowl]

60. No he's not

He's claiming what he found in the available data, not that the problem doesn't exist.

I do disagree with you that it's wrong to report findings when you know your dataset might have shortcomings as long as you are honest about it. If you don't then nobody is likely to invest money in producing a better dataset. It's obvious that he's thought long and hard about the issue and looked for a wide variety of explanatory factors. Just because the outcome is surprising doesn't mean it's wrong; maybe there's an explanation that we haven't thought of yet, maybe the data we have at present is inadequate or has systemic bias in its compilation, maybe police are in fact a bit more careful with pulling the trigger than the way they are strereotyped (and I've had comments hidden for anti-police bias so I'm reluctant to include that possibility...but it's be dishonest if I didn't).

In other branches of science you try to clean your data up as much as possible but having done that you go ahead and publish your surprising result. That's happening in the physics world right now, where scientists are trying to verify or refute the apparent detection of a new subatomic particle that is not predicted by the Standard Model of physics and thus has the potential to upset the whole applecart. The scientists who first noticed it checked and rechecked their result and finally published a paper along the lines of 'this is probably an equipment failure but we haven't managed to find one, and if it is true then it would be a big deal.' Then scientists at a different facility published a paper along the lines of 'yeah we found that too using totally different equipment so that's interesting.' And now they're working on doing experiments at the Large Hadron Collider that should give a definitive answer.

If the original scientists hadn't published their inexplicable result then the matter would go un-investigated because it was so far outside what people expected. Same thing here.