Bootstrapping Surveys

Here’s an interesting epistemological news item. Don’t know all that much about it, but here’s what I know. There’s a worry that telephone surveys are unreliable instruments, since it is now fairly common to screen calls and much more common to refuse to talk to people you don’t know on the phone than it used to be. So, it appears that the people who respond to such surveys over the phone may be strange in important respects. For example, maybe they are lonely, or have little to do with their time.

As a result, there’s an empirical question to be answered about the bias involved in phone surveys. Turns out certain groups of people are more likely to respond to a telephone survey than other groups of people, leading to various attempts to shore up the reliability of sampling techniques.

Here’s the kicker, though: they confirmed the bias of phone surveys by conducting a phone survey…!

So, now when asked whether I believe in bootstrapping, I can reply, “Believe in it? I’ve seen it!”

Leave a Reply

Your email address will not be published. Required fields are marked *