As you ponder the presidential election results, here are some issues to consider. Trump outperformed his polling predictions in each of his presidential runs. Was the Bradley Effect in play this time?
Although the final vote tallies are close between Harris and Trump—less than 2 percent–more people voted for Trump than expected. As pollsters try to figure out what went wrong in their predictions, I will make the same suggestion I made in 2020. How about desirability bias, also known as the Bradley Effect? Desirability bias, which had been around a while, was introduced into election polling discussions to explain why African American Tom Bradley, the popular mayor of Los Angeles, lost his election for governor of California in 1982 when he was leading the polls.
The Bradley Effect proposes that polls are sometimes skewed by social desirability bias. A small percentage of voters may be hesitant to state their real preference because they may perceive that their choice is not publicly acceptable, despite the interviewers’ assurances that their responses will not be related to them individually.
Bradley was an African American running against a white man, and some white voters are thought to have been reluctant to admit that they were voting for the less popular white guy and against the black guy. At the time, pollsters estimated the Bradley effect to be as much as five percent—the actual vote differences from the polls.
The Bradley Effect has been cited in several elections since Bradley’s, including Andrew Gillum’s race for Florida governor in 2018. Gillum was leading in the polls but lost the election to Ron DeSantis.
It is possible that some voters who were not going to vote for a black woman either lied about their intentions when polled or did not vote.
Or was it the shy Trump Voter Effect in conjunction with the Bradley Effect? The shy Trump Effect is based on the so-called shy Tories Effect in Britain, where pollsters were thought to underestimate the potential vote for the conservative Tories because of their social stigma.
Similarly, a shy Trump effect might have been operative because of the possible stigma of a person with all of Trump’s unprecedented and well-known baggage. While many Trump supporters hail their support for him, others may not be willing to go public with it. So, they lie to the pollsters.
Before the election, noted pollster Nate Silver cast doubt on the shy Tories Effect, as it was not in play in 2017 when the Tories underperformed and lost their majority.
And what about Obama, who won twice? Former Virginia governor Douglas Wilder suggested in 2012 that the Bradley Effect was not yet dead, noting that in 1989, he won but by a much smaller margin than predicted, as was the case with David Dinkins in his small margin of victory in New York City in the same year. He also pointed to an article in the New York Times in 2012 that proclaimed, “Race is Still an Issue for Some Voters.”
Please note that since the political advent of racist Donald Trump, racism has increased in America. For example, the Anti-Defamation League counted 283 events organized or attended by white supremacists in 2023, a 63 percent increase from the 173 recorded in 2022.
On the other hand, a black candidate has won the presidency, but not a female. Consequently, Harris suffered what might be called the Hillary effect—polling for the woman but voting for the man.
Upholding the established hierarchical social order, as with the Tories, is one thing, but racism is another. It is much stronger, and when joined with misogyny, it presents a solid double whammy.