Posted by Jay Livingston
(This is an update to yesterday’s post about Halloween greed.)
It turns out, there’s a nice Halloween field experiment I was unaware of (Diener, et. al, JPSP, 1976). Here’s the setup. On Halloween, a woman answers the door and invites the trick-or-treaters in. She tells them to “take one,” and then leaves the room leaving the bowl of candy and a bowl of nickels and pennies.* Overall (27 houses, 1300 kids) most kids (69%) took one. But conditions mattered.
In one experimental manipulation, the woman either asked the kids who they were and where they lived or she allowed them to be anonymous. Experimenters also noted whether the kids were trick-or-treating alone or in groups. For some groups, the woman designated the smallest kid in the group as being in charge of making sure that kids took only one. All these variables made a difference.
Those who did cheat usually took only an additional one to three candies. But of those who did grab more than what was offered, 20% took both candy and coins. Unfortunately, the Snickers study is not like the marshmallow study, so we don’t know where those greediest kids are now.
* Adjusting for inflation that would be closer to dimes and quarters, maybe even half-dollars.