A Friday Washington Post article decried a growing gap in racial attitudes between white members of the Republican and Democratic parties. Liberal media lambasted this supposed evidence of growing Republican racism. But numbers the Post’s report ignored show the opposite – racism among Republicans has been plummeting for years. What the Post calls “never bigger” is far less significant and not all that interesting – the margin between attitudes in each party.
Though the raw data from the National Opinion Research Center at the University of Chicago is not easily available on line, the Post helpfully provided me with a spreadsheet. Here’s the trend among white members of each party who say blacks are less intelligent:
- Republicans (1990): 57 percent
- Republicans (2000): 39 percent
- Republicans (2010): 33 percent
- Republicans (2016): 26 percent
- Democrats (1990): 60 percent
- Democrats (2000): 40 percent
- Democrats (2010): 26 percent
- Democrats (2016): 18 percent
But readers of Aaron Blake’s article in “The Fix” would be unaware of those dramatic drops unless they studied the graphs carefully. Blake’s analysis of that question, well, fixated on a ripple in the percentage point differences between the parties:
- Gap between Republicans and Democrats (2004): 6 points
- Gap between Republicans and Democrats (2010): 7 points
- Gap between Republicans and Democrats (2016): 8 points
That’s the news? Single percentage point shifts are rarely statistically significant. And they are dwarfed by the unmentioned rapid evaporation of bigoted attitudes among members of both parties in America.
A survey is only as good as its questions, and in this case several of the questions (written by the original researchers, not the Post) are poor yardsticks for measuring problematic racial attitudes. For example, the question whose results Blake called “the biggest yawning gap” asks if African Americans are worse off economically “because most just don’t have the motivation or will power to pull themselves up out of poverty.” Leave will power aside. Boosting motivation among the poor was the main impulse behind bipartisan (but Republican-led) welfare reform efforts, which tried to solve the problem of people avoiding work to keep their benefits.
Now, maybe some respondents thought the reference to lack of motivation referred to work ethic rather than public policy, but we don’t know because the survey’s language is ambiguous. Given the dual meanings of “motivation,” the discrepancy on that question is unsurprising, because strategies to foster independence among the poor are a significant philosophical difference between the parties.
Then there’s a question Blake says is about affirmative action, though it does not use the term: “What do you think the chances are these days that a white person won’t get a job or promotion while an equally or less qualified black person gets one instead?” When did acknowledging that de facto racial preferences exist become a racially insensitive “view of the intelligence and work ethic of African Americans”? Come on. Republicans are more likely to oppose affirmative action, so of course they think it’s more widespread.
Another question referred to how the respondent would feel if a close relative married a black person. That deeply personal question has nothing to do with public policy, and doesn’t even ask if respondents approve of interracial marriage in general – just how they would feel if it occurred in their own family. Here again, the Post thinks the most important detail in the data is that 14 percent more Republicans than Democrats would be uncomfortable with such an event in their family – omitting the astonishing and very relevant fact that just in 1990 a whopping 67 percent of Republicans were uncomfortable as opposed to 26 percent today.
Given its irresponsible first go-round, Post should revisit the data trends on racist attitudes, and help Americans celebrate the steep decline of problematic attitudes among Republicans and Democrats alike – rather than continue to beat the partisan drum by ginning up statistics that are all trees and no forest.