Download data and study materials from OSF
Principal investigators:
Neil Malhotra
Stanford University
Email: neilm@stanford.edu
Homepage: http://www.stanford.edu/~neilm/
Alexander Kuo
Cornell University
Email: agk72@cornell.edu
Homepage: http://government.arts.cornell.edu/faculty/kuo/
Sample size: 397
Field period: 5/26/2006-5/31/2006
When government fails, whom do citizens blame? Do these assessments rely on biased or content-rich information? Despite the vast literatures on retrospective voting in political science and attribution in psychology, there exists little theory and evidence on how citizens apportion blame among public officials in the wake of government failure. We designed a survey experiment in which respondents ranked seven public officials in order of how much they should be blamed for the property damage and loss of life in New Orleans after Hurricane Katrina. We manipulated the information provided to respondents, with some receiving the officials’ party affiliations, others receiving their job titles, and others receiving both cues. We find that party cues cause individuals to blame officials of the opposite party, but citizens make more principled judgments when provided with information about officials’ responsibilities. These results have implications for our understanding of the impact of heuristics and retrospective evaluations of government performance.
Hypothesis 1: Republicans/Democrats should be more likely to blame Democratic/ Republican officials when provided with party cues.
Hypothesis 2a: When provided with both the party cue and office cue, the party cue should dominate.
Hypothesis 2b: When provided with both the party cue and office cue, the office cue should dominate.
Hypothesis 2c: When provided with both the party cue and office cue, then the effect of the combined cue should be a mixture of the effects of the individual cues.
The respondents were randomly assigned to one of four experimental groups, which differ along two dimensions with respect to the information contained in the response options: whether each public official in the ranking list is associated with his/her office title, and whether the official is given a party affiliation. These two dimensions constitute the four experimental groups. The control group only received the list of seven proper names without any additional information. Group 2, the “party cues” condition, received the list of proper names with each official’s partisan affiliation. Group 3, the “office cues” condition, received the list of proper names with each official’s job title. Group 4, the “both cues” condition, received the list of proper names with each official’s partisan affiliation and job title. For all respondents, the order of the officials on each list of officials was randomized.
Our main dependent variable in the survey is the blame ranking assigned to each public official by the respondent. Respondents were asked: “Who do you think should be blamed the most for the loss of life and property damage in New Orleans that was caused by Hurricane Katrina?,” followed by the list of seven officials. As explained above, the survey response options vary with respect to whether information about party affiliation and/or job title is provided. After selecting an official, respondents were then asked: “Who do you think should be blamed the second most for the loss of life and property damage in New Orleans that was caused by Hurricane Katrina?” A series of six questions was asked until respondents had ranked all seven officials.
We find that party cues significantly influence whom respondents blame for the damage and death caused by Hurricane Katrina. Compared to the control group, receiving the party cue makes Democratic respondents 1.75 times more likely to blame Republican officials. Compared to the control group, receiving the party cue makes Republican respondents about 1.48 times less likely to blame Republican officials. Regarding individual variation in the use of cues, we find that respondents for whom Hurricane Katrina was personally important were less likely to be affected by the party cues. However, when respondents were given both the party cue and office cue (multiple pieces of information), in the vast majority of cases, the office cue either diluted or dominated the effect of the party cue. This suggests that when given such information, respondents took into account the positions of the officials and used that information to form a reasoned opinion (or at least used it to reduce the bias of the party cue).
We demonstrate that citizens of all levels of political sophistication utilize content-rich, relevant information to mitigate partisan bias in determining who is responsible when government actors fail to perform their duties competently. We find that party cues cause individuals to blame officials of the opposite party, but citizens make more principled judgments when provided with information about officials’ responsibilities. These results have implications for our understanding of the impact of heuristics and retrospective evaluations of government performance. We also show that individual characteristics such as the personal importance of the issue make people less reliant on party labels to make informed decisions regarding blame. Our results give us reason to be cautiously optimistic about the capacity of citizens to make unbiased blame attributions, an important responsibility in democratic systems. Although citizens do not appear to be the objective processors of information envisioned, they are also not myopic stooges characterized by recent critiques of the retrospective voting literature. In other words, we find that people do the best with the information they have.
Malhotra, N., and A. G. Kuo. 2008. "Attributing blame: The public's response to Hurricane Katrina." The Journal of Politics 70:120-135.
Malhotra, Neil. 2008. "Partisan Polarization and Blame Attribution in a Federal System: The Case of Hurricane Katrina." Publius: The Journal of Federalism. 38: 651-670.