The Pollster and Popular Racism

The Surprisingly Easy Acceptance of Racism in Canada

Polling firms use some of their polls as a marketing tool. They want some news attention and a poll on politics is usually an easy story to get from any newsroom. All you have to do is write up a few words, stick in some numbers, and away you go.

The number of people working in newsrooms who actually understand polls is pretty small. There are lots of smart people in newsrooms, scary-smart people, but that doesn’t mean they understand polling enough to compare the apples of one poll to the apples of another or pick up obvious nonsense.

Not just newsrooms. Not just polls. In the recent United States Supreme Court decision on affirmative action, Justice Kentanji Brown Jackson dissented. In her written dissent, Brown quoted information contained in brief submitted by the Association of American Medical Colleges in support of affirmative action.

“For high-risk Black newborns, having a Black physician more than doubles the likelihood that the baby will live.” Affirmative action saves lives went the logic. But the math is absurd. As Ted Frank noted in the Wall Street Journal, if 40% of Black infants died, there’d still be a 60% survival rate. It would be impossible to double that since you can never have more than 100% of anything.

As as it stands, while there’s no question that the infant mortality rates for Black infants in the United States is significantly worse than for White infants, 99% survive their first year of life. The wonky math is one thing. Both the AAMC and by extension Justice Brown relied on a 2020 study that looked for any connection between improved infant health and the race of the physician treating the child. Contrary to what the AAMC claimed and Brown repeated, the study found that “Black infants experience inferior health outcomes regardless of who is treating them.” In other words, Brown accepted a claim that aligned with her beliefs despite the fact it was absurd on the face of it and didn’t actually come from the study cited as the source.

With that in mind, consider recent marketing poll by Pollara on federal politics. It’s garnered some public but so far no one has picked up on either of two issues. One is the way Pollara offers racially based poll results that are actually racist in their assumptions. The other is a few serious serious flaws in the poll itself and the way Pollara presented the information. No surprise as humans don’t question what we assume is normal.