Confidence Intervals

Our society seems to basically work on a 95% confidence interval - anything outside two standard deviations from the norm might as well not exist. And I don't just mean that this is the confidence interval that we use when we have to pick confidence intervals. I mean that we think in these terms about pretty much everything, whether we are conscious of it or not. When we make statements about what is or is not, or will or not happen, this is what we actually mean. The 2.5% outside of this doesn't have any meaning or existence - or might as well not. 

Partly this is necessary - building up knowledge is, of course, principally an activity of tolerating the limits of our confidence in that knowledge. But where it become a problem is when we start talking about people. There, our epistemic habits (of working with the 95% interval) clash with our moral commitments to treat all people - or rather each person - with equal concern and respect. Suddenly, the mental tools we're so used to using don't have any way to keep up; they can't deliver what we morally need them too. I'm not sure we've come to grips with this. 

A lot of academic research does, in fact, take place about people outside the confidence interval. Certainly much critical work is precise an attempt to bring them in. But it very much seems to be an exercise of "oh, and what about ____". Until we can manage to overcome or foundations in the 95%, it's going to remain a sideshow.