Search TESS:


Sources of Blame Attribution: Citizen Attitudes Towards Public Officials after 9/11

Download data and study materials


Principal Investigator(s):

Neil Malhotra
Stanford University
Home page:

Alexander Kuo
Cornell University
Home page:

Sample size: 1015
Field period: 2/14/2007


When government fails, whom do citizens blame? Despite the vast literatures on retrospective voting and attribution, there exists little theory and evidence on how citizens apportion blame among multiple public officials in the wake of government failure. We address this omission by proposing a simple survey experiment exploring citizen attitudes regarding blame of intelligence officials for making America vulnerable to the attacks on 9/11. We manipulate the party labels attached to various government officials and then asked respondents to rate and rank these figures in terms of how responsible they should be for failing to prevent the 9/11 attacks. We hypothesize that partisanship should bias how people attribute blame in the wake of government failure. We find that the domain-relevance of the official moderates partisan rationalization.


Hypothesis 1: Switching the labels of domain-relevant officials from Democrat to Republican may cause Republican citizens to blame them less and Democratic citizens to blame them more.

Hypothesis 2: There should be no partisan rationalization when citizens are evaluating officials irrelevant to the domain.

Experimental Manipulations:

We propose a short Internet-based survey experiment to test whether party cues affect people’s attitudes about the responsibility of three government officials in making America vulnerable to the 9/11 attacks: CIA Director George Tenet, FBI Director Louis Freeh, and Federal Reserve Chairman Alan Greenspan (see Appendix 1 for the questionnaire). Our proposed experiment consists of three manipulations, one for each of the three officials. In each manipulation, respondents will be randomly assigned to one of three conditions. In Condition 1, the official will be designated as a “Republican appointee.” In Condition 2, the official will be designated as a “Democratic appointee.” In Condition 3, which serves as the control group, no designation will be provided. Since each of these officials was appointed by both a Republican and Democratic president, the information presented to respondents is correct, but provides differing party cues.

Key Dependent Variables:

Respondents in all groups will be asked to rate the three public officials in terms of how much they should be blamed for making American vulnerable to the 9/11 attacks.

Summary of Findings:

Both hypotheses were confirmed. Democratic respondents blamed Tenet and Freeh (domain-relevant officials) more when the label was switched from Democrat to Republican; Republican respondents exhibited the reverse effect. No comparable effects were observed for Greenspan (an official irrelevant to the domain).


Healy, Andrew, Alexander Kuo, and Neil Malhotra. 2014. “Partisan Bias in Blame Attribution: When Does It Occur?” Journal of Experimental Political Science. 1: 144-158.

Malhotra, Neil. 2009. “Order Effects in Complex and Simple Tasks.” Public Opinion Quarterly. 73: 180-198.

Copyright © 2014, TESS