Scott Adams has just written an article proposing an experiment that I’d like to try. Research has shown that people are significantly more likely to believe something that conflicted with their previous beliefs if they write about it: in the experiments that lead to this discovery, volunteers were asked to write about a viewpoint other than their own on a given issue, and it was observed that within a few months there was a reasonable likelyhood that they had changed their beliefs to those they justified in writing. It turns out that making people read about a point of view is not nearly so effective at persuading them to adopt it as making them write an argument for it does.
This experiment involves standing around somewhere with a clipboard and offering a token reward (a quid, or a chocolate bar, or something) in exchange for participating in a study into handwriting when writing lies (this isn’t actually what we’re doing, but hey…). Participants are asked to write a couple of sentences about how attractive the experimenter is, and, if they’re willing to be contacted about "further research," to leave their e-mail address or phone number (on an appropriately laid-out form).
Some time later, the experimenter will have a list of contact details for people, many of whom will find the experimenter more attractive than they did when they first met. It’s an instant "little black book."