Bradley effect

From Wikipedia, the free encyclopedia

The term Bradley effect or Wilder effect refers to an explanation advanced as the possible cause of a phenomenon which has led to inaccurate voter opinion polls in some American political campaigns between a white candidate and a non-white candidate.^[1][2][3] Specifically, there have been instances in which such elections have seen the non-white candidate significantly underperform with respect to the results predicted by pre-election polls. Researchers who have studied the issue theorize that some white voters gave inaccurate polling responses because of a fear that by stating their true preference, they might appear to others to be racially prejudiced.

The theory suggests that statistically significant numbers of white voters tell pollsters in advance of an election that they are either undecided, or likely to vote for the non-white candidate, but that those voters exhibit a different behavior when actually casting their ballots. White voters who said that they were undecided break in statistically large numbers toward the white candidate, and many of the white voters who said that they were likely to vote for the non-white candidate ultimately cast their ballot for the white candidate. This reluctance to give accurate polling answers has sometimes extended to post-election exit polls as well.

Some research has suggested that the race of the pollster conducting the interview may factor into that concern. Meanwhile, some pollsters express doubt altogether that deliberately false answers from white voters being polled has been the cause of the polling errors in question. At least one prominent researcher has suggested that with regard to pre-election polls, the discrepancy can be traced in part by the polls' failure to account for general conservative political leanings among late-deciding voters.

[edit] Origin

L. Douglas Wilder's margin of victory in the 1989 Virginia gubernatorial election was narrower than predicted by pre-election and exit polls.

The name Bradley effect is derived from a 1982 campaign involving Tom Bradley, the long-time mayor of Los Angeles, California. Bradley, who was black, ran as the Democratic party's candidate for Governor of California against Republican candidate George Deukmejian, who was white. The polls on the final days before the election consistently showed Bradley with a lead.^[4] In fact, based on exit polls on election day, a number of media outlets projected a Bradley win that night; early editions of the next day's San Francisco Chronicle featured a headline proclaiming "BRADLEY WIN PROJECTED". However, Bradley narrowly lost the race. Post-election research indicated that a smaller percentage of white voters actually voted for Bradley than polls had predicted, and that voters who had been classified by those polls as "undecided" had gone to Deukmejian in statistically anomalous numbers.^[2]

One month prior to the election, Bill Roberts, Deukmejian's campaign manager at that time, had predicted this behavior. He told reporters that he expected that Deukmejian could advance approximately five percentage points from what his poll numbers indicated, due to white voters giving inaccurate polling responses in order to conceal a racial prejudice. Roberts's comments were disavowed by Deukmejian, and the controversy that surrounded them ultimately led to Roberts's resignation.^[5]

Similar voter behavior was noted in the 1989 race for Governor of Virginia between black Democratic candidate L. Douglas Wilder (right) and white Republican candidate Marshall Coleman. In that race, Wilder prevailed, but by less than half of one percent, despite pre-election poll numbers that showed an average lead for him of nearly nine percent.^[6][7] Again, the discrepancy was widely attributed to white voters who had told pollsters that they backed Wilder, but who did not actually vote for him.^[8] As a result of this race, some re-christened the "Bradley effect" as the "Wilder effect".^[9][10] Both terms are still used, each referring to the same dynamic, and less commonly, the term "Dinkins effect" is also used,^[3] in reference to 1989 election of David Dinkins as Mayor of New York City over Rudy Giuliani.

[edit] Other instances

In 1980s campaigns against white candidates, Harold Washington (left), Jesse Jackson (center) and David Dinkins (right) each showed stronger support in the polls than they ultimately received in the voting booth.

Other races which have been cited as possible demonstrations of the Bradley effect include the 1983 race for Mayor of Chicago, the 1988 Democratic primary race in Wisconsin for President of the United States, and the 1989 race for Mayor of New York City.^[7][10][11]

The 1983 race in Chicago featured black candidate Harold Washington running against white candidate Bernard Epton. More so than the California governor's race the year before,^[12] the Washington-Epton matchup evinced strong and overt racial overtones throughout the campaign.^[13][14] Two polls conducted approximately two weeks before the election showed Washington with a 14-point lead in the race. A third conducted just three days before the election confirmed Washington continuing to hold a lead of 14 points. But in the election's final results, Washington won by less than four points.^[7]

In the 1988 Democratic presidential primary in Wisconsin, pre-election polls pegged black candidate Jesse Jackson — at the time, a legitimate challenger to white candidate and frontrunner Michael Dukakis — as likely to receive approximately one-third of the white vote.^[15] Ultimately, however, Jackson carried only about one quarter of that vote, with the discrepancy in the heavily white state contributing to a large margin of victory for Dukakis over the second-place Jackson.^[16]

In the 1989 race for Mayor of New York, a poll conducted just over a week before the election showed black candidate David Dinkins holding an 18-point lead over white candidate Rudy Giuliani. Four days before the election, a new poll showed that lead to have shrunk, but still standing at 14 points. On the day of the election, Dinkins prevailed by only two points.^[7]

Also sometimes mentioned are the 1990 Senate race in North Carolina between black candidate Harvey Gantt and white candidate Jesse Helms, the 1991 race for Mayor of the City of Houston between Texas State Representative Sylvester Turner and Bob Lanier (Turner, who is African American, was the leading candidate to become Houston's first black mayor where an investigative news story pertaining to insurance fraud derailed his first campaign; former Houston Police chief Lee P. Brown (first African American police chief in the City of Houston appointed by Mayor Kathy Whitmire) ran successfully in late 1997. Turner ran again in 2003 finishing third where Bill White won the runoff election), and the 1992 Senate race in Illinois between black candidate Carol Moseley Braun and white candidate Richard Williamson. Gantt lost his race by six points. Two late polls showed Gantt ahead by four to six points, but one other showed a four-point Helms victory.^[17][7] Braun won her general election race by 10 points, but polls indicated a margin of up to 20 points. However, polls had been just as erroneous, though this time underestimating Braun's support, during the primary election. Braun won that contest — also against a white candidate — by three points after polls predicted she would lose by double digits.^[7]

A few analysts, such as political commentator and The Weekly Standard editor Fred Barnes, attributed the four-point loss by Indian-American candidate Bobby Jindal in the 2003 Louisiana Governmental runoff election to the Bradley effect. In making his argument, Barnes mentioned polls that had shown Jindal with a lead.^[18] Others, such as National Review contributor Rod Dreher, countered that later polls taken just before the election correctly showed that lead to have evaporated, and reported the candidates to be statistically tied.^[19][20] In 2007, Jindal ran again, this time securing an easy victory, with his final vote total^[21] remaining in line with or stronger than the predictions of the polls conducted shortly before the election.^[22]

Colin Powell was reportedly warned of the Bradley effect when he was considered to be a potential 1996 presidential candidate.

Inaccurate polling statistics attributed to the Bradley effect have not always been limited to pre-election polls. In the initial hours after voting concluded in the Bradley-Deukmejian race in 1982, similarly inaccurate exit polls led some news organizations to project Bradley to have won.^[23] Exit polls in the Wilder-Coleman race in 1989 also proved inaccurate in their projection of a ten-point win for Wilder, despite those same exit polls accurately predicting other statewide races.^[24][7][6] In 2006, a ballot measure in Michigan to end affirmative action generated exit poll numbers showing the race to be too close to call. Ultimately, the measure passed by a wide margin.^[25]

In 1995, when Colin Powell's name was floated as a possible 1996 presidential candidate, Powell reportedly spoke of being cautioned by publisher Earl G. Graves about the phenomenon described by the Bradley effect. With regard to opinion polls showing Powell leading a hypothetical race with then-incumbent Bill Clinton, Powell was quoted as saying, "Every time I see Earl Graves, he says, 'Look, man, don't let them hand you no crap. When [white voters] go in that booth, they ain't going to vote for you.'"^[26][10]

Ironically, a similar phenomenon was noticed during the early 1990's electoral contests with former Ku Klux Klan leader and Nazi sympathizer David Duke. Many potential voters would not tell pollsters that they favored Duke (as they feared the ostracization that could result from being on record as being a Duke supporter), but would go on to vote for him anyway. The commentary at that time was that Duke "flies under the radar."

[edit] Causes

The issue of which specific sets of circumstances have led to polling inaccuracies is debated, but a general belief among pollsters is that perceived societal pressures have led some white voters to be less than forthcoming in their poll responses. These voters supposedly have harbored a concern that declaring their support for a white candidate over a non-white candidate will create a perception that the voter is racially prejudiced.^[27][28] During the 1988 Jackson presidential campaign, Murray Edelman, a veteran election poll analyst for news organizations and a former president of the American Association for Public Opinion Research, found the race of the pollster conducting the interview to be a factor in the discrepancy. Edelman's research showed white voters to be more likely to indicate support for Jackson when asked by a black interviewer than when asked by a white interviewer.^[3] Andrew Kohut, who was the president of the Gallup Organization during the 1989 Dinkins/Giulianai race and later president of the Pew Research Center, which conducted research into the phenomenon, has suggested that the discrepancies may arise, not from white participants giving false answers, but rather from white voters who have negative opinions of blacks being less likely to participate in polling at all than white voters who do not share such negative sentiments with regard to blacks.^[29]

While there is widespread belief in a racial component as at least a partial explanation for the polling inaccuracies in the elections in question, it is not universally accepted that this is the primary factor. Peter Brodnitz, a pollster and contributor to the The Polling Report newsletter, worked on the 2006 campaign of black U.S. Senate candidate Harold Ford, Jr., and Brodnitz indicated that he did not find the race of the interviewer to be a factor in voter responses in pre-election polls. Brodnitz suggested that late-deciding voters tend to have moderate-to-conservative political opinions and that this may account in part for last-minute decision-makers breaking largely away from black candidates, who have generally been more liberal than their white opponents in the elections in question.^[3] Additionally, with regard to the 1982 contest between Bradley and Deukmejian, Mark DiCamillo, Director of the The Field Poll, which was among those that had shown Bradley with a strong lead, said that the organization's own internal examination after that election identified other possible factors that may have contributed to their error.^[30]

One of the more ardent critics of the acceptance of the Bradley effect as an accurate explanation for observed polling errors is Gary Langer, who serves as the director of polling for ABC News. Langer has described the Bradley effect as "a theory in search of data." He has argued that inconsistency of its appearance, particularly in more recent elections, casts doubt upon its validity as a theory.^[31]

[edit] Diminished effect

Polling numbers in Harold Ford's 2006 U.S. Senate campaign did not exhibit the Bradley effect.

In 2006, there was speculation that the Bradley effect might appear in the Tennessee race for United States Senator between Harold Ford, Jr. and white candidate Bob Corker.^{[32][33][17][27][10]} Ford lost by a slim margin, but an examination of exit polling data indicated that the percentage of white voters who voted for him remained close to the percentage that indicated they would do so in polls conducted prior to the election.^[10][34] Several other 2006 biracial contests saw pre-election polls predict their respective elections' final results with similar accuracy.^[7]

One exception was in the race for United States Senator from Maryland, where black Republican candidate Michael Steele lost by a wider margin than predicted by late polls. However, those polls correctly predicted Steele's numbers, with the discrepancy in his margin of defeat resulting from their underestimating the numbers for his white Democratic opponent. Those same polls also underestimated the Democratic candidate in the state's race for governor — a race in which both candidates were white.^[7]

The overall accuracy of the polling data from the 2006 elections was cited, both by those who argue that the Bradley effect has diminished in American politics,^[27][7] and those who doubt its existence in the first place.^[31] When asked about the issue in 2007, Douglas Wilder indicated that while he believed there was still a need for black candidates to be wary of polls, he felt that voters were displaying "more openness" in their polling responses and becoming "less resistant" to giving an accurate answer than was the case at the time of his gubernatorial election.^[35]

[edit] Barack Obama and the "reverse" Bradley effect

Some have suggested that
Senator Barack Obama may have encountered both the Bradley effect, and a "reverse" Bradley effect, during the 2008 Democratic presidential primary elections.

The 2008 presidential campaign of Barack Obama, a black United States Senator, has brought a heightened level of scrutiny to the Bradley effect theory, as observers searched for signs of the effect in comparing Obama's polling numbers to the actual election results during the Democratic primary elections.^{[10][34][3][36][37]} After a victorious showing in the Iowa caucuses, where votes were cast publicly, polls predicted that Obama would also capture the New Hampshire Democratic primary election by several percentage points over Hillary Clinton, a white senator. However, Clinton defeated Obama by three points in the New Hampshire race, where ballots were cast secretly, immediately initiating suggestions by some analysts that the Bradley effect may have been at work.^[38][39] However, other analysts cast doubt on that theory, offering alternative explanations and pointing to the fact that the polls underestimated Clinton rather than overestimating Obama.^[40]

After the Super Tuesday elections of February 5, political science researchers from the University of Washington found trends suggesting the possibility that with regard to Obama, the effect's presence or absence may be dependent on the percentage of the electorate that is black. The researchers noted that to that point in the election season, opinion polls taken just prior to an election tended to overestimate Obama in states with a black population below eight percent, to track him within the polls' margins of error in states with a black population between ten and twenty percent, and to underestimate him in states with a black population exceeding twenty-five percent. The first finding suggested the possibility of the Bradley effect, while the last finding suggested the possibility of a "reverse" Bradley effect in which black voters might have been reluctant to declare to pollsters their support for Obama. By comparison, with only one exception, in each state with inaccurate opinion polls for the Democratic contest involving Obama, those same polls accurately predicted the outcome of that state's Republican contest, featuring only white candidates.^[41]

While their cause continues to be debated, the pollsters' errors have raised expectations that as the presidential primary season progresses, Obama's polling numbers will be widely scrutinized as analysts try to definitively determine whether the Bradley effect has become a significant factor in the race.^[42]

[edit] See also

• Social desirability bias
• Flora MacDonald, candidate for leadership of the Progressive Conservative Party of Canada in 1976, whose worse-than-expected showing gave rise to the term "Flora Syndrome," positing a similar effect for a female candidate

[edit] Notes