Search TESS:


A randomized experiment on the determinants of generosity to the poor

Download data and study materials


Principal Investigator(s):

Christina M. Fong
Carnegie Mellon University
Home page:

Robyn Dawes
Carnegie Mellon University

Erzo Luttmer
Dartmouth College
Home page:

Sample size: 523
Field period: 9/8/2006 - 9/20/2006




Experimental Manipulations:

We conducted dictator games in which respondents decided how much, if any, of a $100 pie to allocate to recipients of two food pantries in Tuscaloosa Alabama. We implemented 10% of the games with real money, so the average stake size was $10.00. We used a short audiovisual presentation to manipulate perceptions about the race, deservingness, political ID, and religiosity of the recipients. We also provided some respondents with the information that Tuscaloosa is relatively well off compared to the rest of Alabama. We used photos to manipulate race by showing mostly black recipients to some respondents and mostly white recipients to other respondents. We used accompanying audio information for the manipulations of deservingness, political ID, and religiosity or recipients, and Tuscaloosa┬┤s aggregate income.

Key Dependent Variables:

Our primary dependent variable is offers made in the dictator game. We also collected measures of beliefs about recipients, various self-reported measures of generosity to the poor, and various other attitudinal measures.

Summary of Findings:





Fong, Christina and Erzo Luttmer. 2009. "What Determines Giving to Hurricane Katrina Victims? Experimental Evidence on Racial Group Loyalty." American Economic Journal: Applied Economics, 1: 64-87

Copyright ┬ę 2014, TESS