By Chris | April 21, 2008
This last weekend my university campus was open to prospective students and their families. Following the spring football game, the university hosted a small carnival with amusement rides and traditional carnival games. I went with a couple friends who satisfied their inner thrill-seeker with a ride on the ferris wheel. However, I was able to convince my friend Laura to accompany me on The Octopus, a traditional ride that I had enjoyed at the small amusement park in my hometown. She hated it, screaming incessantly during the ride and appearing nauseous when it was over. So I asked:
“Why did you agree to ride it if you knew you would hate it so much?”
“I didn’t think it would be so scary. The people on the ride earlier didn’t scream at all, so I figured it wouldn’t be too bad.”
“That’s the mistake of selection bias!” I exclaimed, thrilled at this illustration of statistics. The riders we had been watching earlier were not a random sample of the population. They were much more likely to be the type of person who enjoys being spun around in circles at fast speeds.
On another note, all of the traditional games at the carnival were purchased with tokens. You paid $1 per token, and then paid 1-2 tokens to try your luck at winning a stuffed animal. Behavioral economists note that removing cash from a transaction changes consumer behavior. I couldn’t help but think that it is easier to fork over 2 tokens to play rigged games for cheap prizes than it is to hand over actual cash. Tokens act as a buffer that prevents carnival goers from thinking of the game as one would rationally evaluate an investment. Is a 25% chance at a $2 stuffed bird really worth the price of $2? Tokens may also be used because they limit employee theft and discourage people from thinking at the margin.
For the record, I shot baskets at one of the booths and missed both of my shots.