The recent violence in Charlottesville, Va., amplified an ongoing struggle in America about who experiences discrimination and to what extent. Many of the white nationalists who rallied in Charlottesville, for example, feel that white people are discriminated against as much as, or more than, minority groups.
Questioning others' experience of discrimination isn't limited to fringe protest groups. Perceptions of discrimination vary heavily across the U.S. population as a whole, as a June study from the Public Religion Research Institute showed. And those differences tend to fall along partisan lines.
The survey found that a plurality of Americans — 42 percent — perceive "a lot of discrimination" against three groups: black people, immigrants, and gay and lesbian people. But the partisan gap is large: 61 percent of Democrats believed this of all three groups, compared with 19 percent of Republicans.
PRRI broke down the numbers by state. When the states' perceptions of discrimination are lined up against states' votes for Trump in 2016, it shows a clear negative correlation — places where there was bigger perception of discrimination had a lower likelihood of voting for Trump. Reliably liberal California and reliably conservative Wyoming reside at opposite ends of the spectrum.
It's a relatively strong correlation, with an r-value of -0.69 (that's a statistical measure that tells the strength of correlation on a scale of -1 to 1 — a measure closer to 1 or -1 means a strong linear relationship, while a measure closer to zero means a weak linear relationship).
And while states that tend to perceive less of this discrimination also tend to be whiter (85 percent-white Wyoming, for example), and white people also tend to perceive less discrimination against blacks and immigrants than other racial groups do, the white share of a state's population does not correlate to the discrimination data as well as support for Trump does. The r-value between those two series is around -0.44.
The data don't say anything about the direction of correlation (standard journalist disclaimer: "correlation is not causation"), but it's easy to see how this relationship might exist. Trump, after all, made opposing political correctness one of his (literal) rallying cries. Wherever 2016 voters' attitudes about discrimination came from — whether stirred up by Trump or brought on by outside forces (or both) — he certainly took advantage of these feelings.
That said, it's hard to separate Trump from Republicanism in general on this; a similarly strong correlation exists between this discrimination measure and a state's share of Republican voters.
To Robert Jones, the founder and CEO of PRRI, it makes sense for perception of discrimination to be a partisan issue.
"I think that goes to a broader worldview thing of, it fits with a conservative bootstrap theory," he said. " 'If you fail there's no one to blame but yourself.' "
But one PRRI datapoint suggests that something shifted among Republicans between 2015 and 2017. Just two years ago, 46 percent of Republicans believed there was "a lot" of discrimination against blacks. As of this year, that figure was 32 percent. Among independents, however, that figure held steady between those two years (it went from 59 percent in 2015 to 58 in 2017), as it held relatively steady for Democrats (going from 80 percent to 77 percent).
And it's not just PRRI's data. A study on the 2016 presidential election found a "relatively strong indication that racism and sexism were more important in 2016 than they had been in previous elections." The effects were particularly strong on the Republican side, with the impact of racism and sexism (as defined by the researchers) much stronger in 2016 voters' choices than in 2012 or 2008, according to the survey by the University of Massachusetts, Amherst and MacWilliams Sanders Communication.
Meanwhile, a majority of Americans believes race relations have worsened in the past year, an NPR/PBS NewsHour/Marist poll found this week.
Copyright 2020 NPR. To see more, visit https://www.npr.org.
Loading...
