Scott Riding and Dr. J. Quin Monson, Political Science
During the course of last year’s U.S. Presidential Election, many political analysts speculated that racial prejudice among American voters would hurt Democratic nominee Barack Obama in his bid for the White House. Candidates Hillary Clinton and Mitt Romney faced similar questions about gender and religious biases against their campaigns. If these biases exist, their presence runs against the grain of the political ideal — that any qualified candidate should have a fair shot at the highest political office in the country. Our research explores the existence and prevalence of these biases in context of the 2008 Democratic and Republican Presidential Primaries.
This aspect of American politics has been studied before. In fact, the popular Gallup Poll frequently asks respondents to rate their willingness to vote for black candidates, women, Mormons, etc.1 Gallup asks:
Between now and the 2008 political conventions, there will be discussion about the qualifications of presidential candidates — their education, age, religion, race, and so on. If your party nominated a generally well-qualified person for president who happened to be black, would you vote for that person?
However, surveys are notorious for understating prejudice. When directly asked about their private biases, many respondents give a socially acceptable answer instead of their true opinion in an effort to save face.2
Our research uses a novel technique to measure and compare racial, gender, and religious biases among the American public in context of a presidential election. It is called a list experiment, or an item count experiment.3 Instead of directly asking every survey respondent about their prejudices, like Gallup, we randomly divide up the respondents into two groups — a control group, and an experimental group. The control group is presented with the following prompt:
Below are four things that sometimes make people angry or upset. After you read all four statements, please indicate how many of them upset you. We do not need to know which ones upset you, just how many.
1) The way gasoline prices keep going up
2) Professional athletes getting million plus dollar salaries.
3) Requiring that seat belts be used when driving.
4) Large corporations polluting the environment.
The experimental group receives the same prompt, but receives five statements instead of four. The fifth statement is something like, “A woman serving as President.” By calculating the average number of statements from both groups, and then taking the difference between them, we can estimate what proportion of the experimental group objected to the fifth statement. In this way we can indirectly measure bias without tipping off the respondent to our intent.
We performed this experiment using eight prompts: 1) a black person, 2) Barack Obama, 3) a woman, 4) Hillary Clinton, 5) a Mormon, 6) Mitt Romney, 7) a Baptist, and 8) Mike Huckabee. The survey was fielded in June 2008 in partnership with Harris Interactive, during the height of the primary campaign. The following graph illustrates the results:
Figure: Estimated percentage of respondents made “angry or upset” by the hypothetical President. Categories with stars next to their description have a statistically significant result.
When asked about a woman or a Mormon, respondents reported significant bias (25% and 27%, respectively). However, asking about a black person or a Baptist results in lower bias and the results are not statistically significant.
These data suggest two conclusions: 1) Americans have come a long way in their acceptance of black candidates, and 2) we still have a journey ahead of us for women and Mormon candidates.
Working with Dr. Monson on this research has been an invaluable component of my education at Brigham Young University. It was an opportunity to take classroom theories and methods and apply them to a timely and relevant question. This was my first direct experience with the ins and outs of academic inquiry and I found the peer review and revision process fascinating. None of this would have been possible without the generous assistance of ORCA.
More than any other aspect of this project, I value the mentoring relationship between Dr. Monson and myself. He helped me understand the scholarly landscape, directed my literature review, and built my confidence as a researcher. We co-authored a paper detailing the results and presented it at three conferences: the PAPOR conference in December 2008, the Southern Illinois conference on Religion and Politics in April 2009, and the prestigious Ohio State conference on the 2008 Election in October 2009. We are currently revising a draft for immediate submission to peer-reviewed journals in our field.