The “Bradley effect” is a polling phenomenon involving high support for non-white and non-female candidates in opinion polls not reflected by election results.
Origins and History
This phenomenon was coined following Tom Bradley’s (D) run for California governor in 1982. Bradley, an African American and the mayor of Los Angeles, was ahead of opponent George Deukmejian (R) entering the final days of the election. On Election Night, Deukmejian defeated Bradley by less than 2% of the vote. Political observers posited that some white voters voiced their support for Bradley in phone polling to avoid appearing politically incorrect or racist.
Additional examples of the Bradley effect followed the namesake’s narrow loss. Chicago mayoral candidate Harold Washington won a more narrow victory over a white opponent than polling indicated. Jesse Jackson’s 1984 and 1988 Democratic presidential campaign received a smaller vote share from white voters than their expectations from polls. Colin Powell considered a Republican presidential run in 1996 but was advised that the Bradley effect would be prominent in an increasingly white party.
The Bradley effect has weakened over time thanks to polling precision and changing cultural values. U.S. Senate candidate Harold Ford (D) received roughly the same amount of white support in polling as he received on Election Night. Barack Obama (D) flipped the Bradley effect on its head in the 2008 presidential election by winning white voters in previously Republican states like Indiana and North Carolina.
FiveThirtyEight’s study of the 2008 Democratic primary between Obama and Hillary Clinton showed the signs of a reverse Bradley effect. Obama received 3.3% more of the vote across all primaries than was indicated by polling including a higher proportion of the white vote. The publication acknowledged that hidden racial motivates may have existed but did not impact Obama’s chances in the aggregate.
The concept of the Bradley effect has been called different names and expanded in scope since 1982. Virginia gubernatorial candidate Douglas Wilder (D) won narrowly in 1989, leading state observers to call it the Wilder effect. David Dinkins (D) won one term as mayor of New York City and lost to Rudy Giuliani in 1993 with some attribution to a Dinkins effect. Hillary Clinton’s 2016 loss to Donald Trump has been attributed to something similar to the Bradley effect. Pew Research Center noted that polls may have underestimated Trump’s support and overestimated Clinton’s support due to concerns by respondents about their level of support.
Vanity Fair (November 3, 2016): “A more significant Bradley effect was visible among certain demographic groups, however. Morning Consult found that voters with a college degree supported Clinton by a 21-point margin in phone interviews, but only by a 7-point margin online.”
The New Republic (October 12, 2008): “But now Lance Tarrance, the pollster for Bradley in that race, has an article up at RCP suggesting that the Bradley Effect was merely a case of bad polling — and that his campaign’s internals had shown a dead heat.”
Politico (October 9, 2008): “There was, in the primary, clearly a ‘reverse Bradley Effect’ among black voters, whose support for Obama was consistently understated in the polling.”