“The Role of Attribution in Accountability: A Natural Experiment Using Train Delays” with Justin de Benedictis-Kessner, MIT

Event time: 
Thursday, October 6, 2016 - 5:30pm through 6:30pm
Location: 
Institution for Social and Policy Studies (ISPS), Room A001
77 Prospect St.
New Haven, CT 06511
Speaker: 
Justin de Benedictis-Kessner, Ph.D. Candidate in American Politics, Massachusetts Institute of Technology
Event description: 

ISPS EXPERIMENTS WORKSHOP

Do people hold politicians accountable for the performance of government? I test this question in a natural experiment using a novel data source: real-world experiences with the performance of a large urban transportation authority. I compile administrative records of service performance tracked at the individual level, link these data to surveys of people’s opinions, and further test the importance of the attribution of responsibility on people’s ability to make retrospective judgments in a survey experiment. I show that people do perceive different levels of performance, but that people fail to connect performance with judgments of government. I find that when people are experimentally provided with information on responsibilities, they are able to connect their own experiences of performance with their opinions of government. However, these judgments are susceptible to a cognitive bias favoring recent information, stunting the ability of citizens to reflect on cumulative performance over time. Broadly, these results demonstrate that confusion about government responsibilities and inherent human perceptual biases can frustrate accountability.

Speaker: Justin de Benedictis-Kessner is a PhD candidate in American Politics at MIT with an interest in political behavior, public policy, urban politics, and methodology. He graduated Phi Beta Kappa from the College of William and Mary in Virginia with a degree in Government and Psychology. His dissertation research examines how citizens hold government accountable, and how both psychology and electoral institutions can subvert their ability to do so. He addresses this question by applying creative data on local politics. His other work examines the role of partisanship in municipal policymaking, statistical techniques for estimating racially polarized voting, partisan media polarization, political campaign strategies, and local political engagement.

Open to: 
Yale Community Only
Admission: 
Free
Event type 
Seminar, Workshop