16/09/2008

The IQ Game

I've long guessed that I could accurately assess a person's IQ plus/minus ten points simply by talking to him or her about the weather or the latest footie results for five minutes. I've also long realized that this guess is simply based on, well, guessing.

Here's a psychology experiment I'd like to see: Take a bunch of undergraduates and randomly assign them to an "assesser" or "assessee" role. Have all of them take IQ tests. Make sure that the assessers (great word!) understand the IQ concept. Have them talk to the assessees (an even better word!) about trivial matters for five minutes; highly g-loaded questions are not allowed. And then let the assessers assess the IQ of the assessees. I'd really like to know how high the correlations are.

If you want to get this kind of research published, you'll have to vary something. No problem: Are the estimates higher if the assessees wear glasses? Or a suit and a tie vs. a jumper and jeans? (Predictions: Yes and yes) Are the estimates higher when the conversations are held in a lab vs. a pub? (Yes) You could also look into the effect of different gender combinations and the assessers own intelligence. (Holding IQ constant, women might be rated lower; the higher the assessers own intelligence, the better the estimate could be.) And if you want to get really wild and have a liberal IRB board, you could vary the assessees level of drunkenness; the more drunk they are, the less accurate the estimate should be: At one point, everyone becomes an idiot.

Speaking of alcohol, this could actually be a fun game if you have a party many guests at which do not know each other beforehand. The shortest IQ tests take only twelve minutes. Make sure everyone arrives on time and bully them into taking one. (You might want to point out that they'll get free drinks.) Hold a short PowerPoint presentation about the IQ concept. Pair them up and let one person estimate the other person's IQ. (Assessers write down a number on a piece of paper, which assessees may not see.) Whoever comes closest wins a copy of The Bell Curve or Dumb and Dumber, depending on your sense of humour.

If the party starts a little slowly, you could use these results, which so far you have kept secret, to spice things up: "Hey, Andy, Suzy has estimated you at 120, whereas actually you're only 95. I think she fancies you." Or conversely: "Hey, Jack, Sam said you're at 80, but the results say you're at 110. Maybe you'll want to have a word with him. Oh, look, there's a knife lying there."

I'll leave the rest to your imagination.

No comments: