> What they miss is that these studies starts with researchers making value judgements, and those judgements aren't more legitimate because a researcher made th

> What they miss is that these studies starts with researchers making value judgements, and those judgements aren't more legitimate because a researcher made them. I disagree. That a researcher (aka expert with significant theoretical and methodological knowledge and experience) makes a judgement on something makes it likely a better judgement than that of a random person. But more importantly I doubt the problem is (usually) the researchers. Most good social science will give confidence intervals, careful language, etc. But how it is then spun and interpreted by university communication departments, news, science bloggers etc is often far from the actual findings.