But ahh, as I found out on HN the other day, it actually is: http://blog.revolutionanalytics.com/2011/04/the-plural-of-an... ;)

Seriously though, after working at a prestigious institution and seeing how often lunch and dinner conversations between peers involved talk of how "my daughter can work in your group!" and "your son can work in my group!", my intuition tells me this is just the way it is - nepotism is pretty much everywhere. Sure, more anecdotes, but that's all I got until someone does some solid research on it.

For now here're two more articles, that may have some more information: http://www.washingtonpost.com/blogs/monkey-cage/wp/2014/01/3...

http://www.theatlantic.com/magazine/archive/2003/07/in-prais...

Actually it is. An anecdote is an observation. And data is just a larger set of observations.

There's no requirement that data must be exhaustive. And in most social science research, it never is (actually that is frequently the case in hard sciences, too).

The issue is that anecdotes come with a whopping big side of selection bias. Sure, it's technically "information", but calling it "data" implies rigor. There are a lot of alien abduction anecdotes, but this tells you more about humans than it does about aliens.