Sigh. It’s time for another “Margaret Wente gets it wrong” blog post.

This past weekend, the Globe and Mail plagiarist columnist proclaimed that racism isn’t really as prevalent as “the progressive left” would have you believe. In support of this position, she cited a study that was recently published by the open-access online journal PLOS ONE. Here’s how Wente introduce the study in her column:

But how prevalent is discrimination, really? U.S. researcher Brian Boutwell and a team of academics set out to answer this question in the most straightforward way. They asked people. They asked more than 14,000 Americans how often they had been discriminated against, and why.

No, they didn’t. The very first sentence in the section of the study headed “Data” states that “Data are derived from the National Longitudinal Study of Adolescent to Adult Health (Add Health)”. In other words, the researchers used data from a secondary source – not data that they collected by asking a question themselves.

There’s nothing wrong with using secondary data in a research study. Secondary data can be an excellent resource, especially for investigations of large-scale social issues like racism, which might need more data than an individual researcher can easily collect. However, the usefulness of secondary data can be limited by how the original survey questions were worded. Boutwell and his colleagues drew their data from responses to two survey questions. This first was: “In your day to day life, how often do you feel you have been treated with less respect or courtesy than other people?” (to be answered with either “never”, “rarely”, “sometimes” or “often”). If the respondent answered “sometimes” or “often”,  then they were asked, “What do you think was the main reason for these experiences?” (to be answered with one of 11 possible responses, including race, ancestry, and skin colour).

The main survey question only asks about experiences of discrimination in one form – experiencing “less respect or courtesy”. Discrimination can also happen in forms such as not getting a callback for a job application because of your having an “ethnic” name, or in being overlooked for a position on a corporate board because of your gender, and in many, many other ways.  It’s completely unreasonable for Wente to claim that racism is “declining”, based on data from a single, very limited, definition of “discrimination”.

Wente then acknowledges that “one study is far from conclusive”, and presents this quote from the conclusion of the Boutwell et al study: “Our results thus provide at least somewhat of a counterweight to possibly exaggerated claims that discrimination is a prevalent feature of contemporary life in the United States.” However, as she has done before when describing research outcomes, she ignores other information in the study that doesn’t fit her narrative.

Two paragraphs below the one that Wente quotes, Boutwell and his colleagues say this:

[M]uch caution is necessary when interpreting our findings. What should be avoided is the conclusion that our results suggest that the problem of discrimination in the US is, to any great extent, remedied and in need of no further scrutiny or improvement. Indeed, the observation of a bivariate association between the racial categories and discrimination experiences…suggests that such experiences vary by race, which should remain a topic of focus for social and behavioral scientists, as well as for members of the community (although perhaps not in ways hitherto suspected).

The researchers explicitly warn that the results of this study should not be used as a reason to make the type of “less racism” claims that Wente is making. But this warning is conveniently not mentioned in Wente’s column.

Wente then further discredits herself by describing one incident of negative responses to racism, and saying “This story proves that everyday racism is not tolerated by ordinary people.”  So, apparently, a single study is far from conclusive, but a single incident is “proof”. What???

Wente is entitled to her opinions. But the Globe and Mail’s readers are entitled to thoughtful analysis based on accurately reported source material. This column isn’t thoughtful, and it isn’t accurate.