Comments on: Help Me Today, I’ll Help You Tomorrow

By: johntemple

johntemple — Mon, 01 Aug 2011 13:04:19 +0000

Very interesting study. In the explanation, you mention only 2 conditions (able to reciprocate giving and unable to reciprocate giving), but it would be interesting to see what happened when you varied the amount of money offered to each player. Specifically, giving player A significantly more money and observing whether he gives any to player B, out of kindness, altruism, or guilt for receiving more money.

By: Merve

Merve — Sun, 03 Jul 2011 23:16:36 +0000

In reply to georgekhasin.

As to your last question, there might be many reasons: They might think that they should give some of the money to the other because this is what’s appropriate (following norms), they might be altruistic – i..e. they might drive utility by giving money to someone else per se. They might feel guilty because they got some money from the experimenter but the other didn’t…. Reasons might be plenty – and controversial. But it is actually a stylized fact that Player As do send positive amounts to Player Bs in dictator games, independent of the money at stake and different cultures. For more info Colin Camerer’s Behavioral Economics is a great source..
Merve

By: georgekhasin

georgekhasin — Thu, 30 Jun 2011 03:52:30 +0000

Very interesting. But it seems to me that this is an example for studies that are not really representative of the general population if they stand alone without two or three non-receptive studies – that is, studies that do not engage the subjects directly.

I believe that the transition of the money in the lab might have some “priming effect” on the participants which lead them to make decisions different than their representative population.

Also, why would player A give any of his money to player B in the second version in which he can not get anything in return? what were the rules of the game?

By: sevgi

sevgi — Sat, 25 Jun 2011 19:04:42 +0000

Hello, thanks a lot for the comment. These results are from an extensive survey (more than 1,000 people), but we can’t tell if people would actually behave in line with what they report. We are actually hoping to run a controlled lab experiment with actual money.

By: Cetin Kaya Koc

Cetin Kaya Koc — Sat, 25 Jun 2011 16:20:31 +0000

You can think of the two-player game as a social experiment and run it (among your friends etc, or if you can get the funding as a controlled experiment). However, it would also be instructive to do this mathematically, i.e., by a Monte Carlo simulation. Set this up as a Monte Carlo experiment and keep running hundreds of thousands of times to see whether “on the average” what results are obtained (based on assumoptions made initially).