It is pretty easy to determine things (with a high degree of confidence) about a large group of people having only encountered a small group of people. Of cour

It is pretty easy to determine things (with a high degree of confidence) about a large group of people having only encountered a small group of people. Of course how you sample is vital. With random sampling, it is pretty straightforward Stats 101 stuff. This wasn't random sampling which throws a wrench in it, but neither are these individual people actually independent. They were all selected by the organization they belong to.