More than 53,000 Black men joined a virtual summit, Win With Black Men, on Monday night to rally behind presumptive Democratic presidential nominee Kamala Harris. During the four-hour conversation, organizers said the group raised more than $1.3 million for Harris’ campaign and grassroots voter advocacy groups focused on Black men.
The success of the conversation, which was inspired by the Win With Black Women conversation the night before, runs counter to the narrative shaped by recent polls showing that 30% of black men plan to vote for Donald Trump. “Don’t let anyone stop us by asking the question, ‘Can a black woman be elected president of the United States?’” Raphael Warnock, the Georgia representative in the U.S. Senate, said on the call. “Kamala Harris can win. We just have to show up. History is watching us, and the future is waiting for us.”
Black voters have historically been a key voting bloc for Democrats, but experts say inaccurate polling of black men in particular could create false narratives about their preferences this election cycle, particularly the idea that there’s been a massive shift of black voters to the Republican Party. Win With Black Men, hosted by journalist Roland Martin, said it works to dispel stereotypes about changes in black male voting behavior, their refusal to support a female candidate and their reluctance to mobilize politically.
“People are making a lot of guesses without actually talking to a large enough group of black people to be able to say things with the precision that they say them,” said Andra Gillespie, an associate professor of political science at Emory University. “You’re not going to be able to detect what’s probably not going to be [than] a one- to three-point shift in favor of Donald Trump, based on changes in surveys where you talk to 200 black people at a time. I can’t say with any statistical certainty whether that three-point shift is real or not.”
Unrepresentative polls can also have a negative effect on voting behavior. People tend to vote when they think the election is close, Gillespie said. So polls that suggest Trump will win easily and that even black people will vote for him in droves can distort people’s understanding of reality. It’s important to make sure the public is aware of potential inaccuracies in polls.
“These stories are also used to confuse Black voters themselves, which in turn can depress the Black vote and lower turnout,” said Christopher Towler, founder of the Black Voter Project (BVP), a national polling initiative. “It can be used as a mechanism to turn off voters, knowing that Black voters are going to play a key role in this election.”
Gillespie said it will be a few days before new polls look specifically at Harris’ performance among black voters. Still, the recent mobilization around Harris suggests that talk of a possible exodus of that bloc from the Democratic Party may have been premature.
The problem comes down to sample size. In surveys of 1,000 to 1,500 voters, subsample sizes of black voters can range from 150 to 300. In some cases, all people of color are lumped together into one demographic group. Surveys with such small sample sizes create large margins of error.
“The problem is the level of precision with which we can make certain statements when you’re talking to so few people,” Gillespie said. “The number that comes out of the survey is the middle of a range of possible numbers that we think are in the real population because of statistical analysis.”
If the subsample size is less than 100, she said, the margin of error is plus or minus 10. So if a poll says 20% of black voters will vote for Trump, the actual number based on the subsample is between 10% and 30%.
Towler, of the Black Voter Project, said he began noticing the problem of unrepresentative polls years ago. He started BVP to “counter the industry standard of including a few hundred black responses to a general survey” and use that small sample size as a complete picture.
“It’s really unscientific,” Towler said. “So I’ve worked for years to try to create data that is reliable, accurate, and truly representative of the black community.”
This year, BVP released a large, multiwave, national poll aimed at collecting representative data from black Americans. The survey was conducted from March 29 to April 18, with 2,004 black Americans interviewed, and collected a nationally representative sample of respondents from all 50 states. The BVP survey found that 15% of respondents would vote for Trump if the election had been held at the time of the survey, a figure that is much lower than reported by other polls with smaller sample sizes. A survey on the BVP scale is important for getting a sense of what the black population and the black voting age population in the U.S. actually looks like, Gillespie said.
But mainstream Beltway polling firms typically lack black leaders, so accurate polls of black communities aren’t a priority, Towler said. What’s more, some pollsters don’t see the value in spending more money to survey a demographic they estimate will already vote staunchly Democratic.
Black and brown communities are on the margins of American politics, says Emmitt Riley, president of the National Conference of Black Political Scientists. The biases of mainstream political scientists result in pollsters that fail to adequately capture the political behavior and positions of marginalized groups.
“A lot of people who study race are not considered mainstream political scholars,” he said. “That has huge implications for the kind of reporting that happens, the kind of news stories that come out, how you describe what’s happening in these communities.”
Towler said pollsters need to create surveys that are culturally competent and ask questions in ways that don’t produce misleading opinions. “It’s important when you’re looking at polls of black people, first of all, not only to make sure that you have polls that are designed to accurately measure black people’s opinions, but also to have pollsters who study and understand the black community,” he said.
While polls are still trying to figure out where black men will throw their political support, groups like Win With Black Men and Black Men for Harris are making their loyalties clear.
“Let’s protect Kamala. Let’s be with her as if she was there for us,” Bakari Sellers, the former South Carolina Rep., said during the call. “We’re going to disagree a lot. But let’s put the petty squabbles aside. Let’s stand up and be the black men who change this country. We built this country. I’m rocking with Kamala.”