Elite Field Experiments to Explore Candidates Accountability and Bias
In an ongoing project that goes back to my Ph.D. dissertation in economics (Bologna University 2006), I analyze candidates’ behavior during the electoral campaign in representative democracies. The original field experiment, inspired by Putnam observational study on the responsiveness of Italian region, sent emails to candidates in the elections, to study if candidates were more responsive to decided or undecided voters. The seminal paper was one of the first to apply email experiment in political science.
Since then email experiments have become fairly common. In 2010 together with Holger Kern, Feliciano Guimaraes and Gabriel Cepaluni we modified the design to test discrimination toward low and high class citizens and native Brazilians (11000 emails). This experiment found no trace of electoral targeting, but identified a significant negative discrimination against native Brazilians, and positive discrimination toward women at the state level, but not at the federal level. Additionally this design employed two rounds of experiments, one before the election and one after the elections, to isolate the effect of the electoral campaign on discrimination.
In October 2012 we implemented two new replications, the first mainly focused on discrimination, the second one focused on clientelism (~40000 emails).
The first Brazilian pilot, conducted in 2008, was published in the Brazilian Political Science Review.“Investigating Elite Behavior Through Field Experiments: do candidates answer more to core or swing voters?”. Joint
with Prof. F. Guimarães, Brazilian Political Science Review, v7(1) 2013. The 2010 panel-experiment has recently received a revise and resubmit from the Journal of Politics.