To keep Whistling Vivaldi readable, Steele intersperses his recounting of academic studies with personal anecdotes of his own or of students and professors he’s met throughout his academic career. * For example, do guidance counselors or academic advisors tell female students not to major in math or engineering? Do they discourage black students from applying to the best possible colleges to which they might gain admission? Marge and Homer may not have taken Lisa’s concerns about Malibu Stacy seriously, but she was more right than even the Simpsons writers of the time (who were probably almost all white men) realized. Stereotype threat can be found, again through experiment, in kids as young as six years old. And Steele argues, with evidence, that stereotype threat begins much earlier than most people aware of the phenomenon would guess. If stereotype threat is directly* or indirectly convincing women not to choose STEM majors, or steering African-American students away from more academically rigorous majors or schools, then we need policy changes to try to address the threat and either throttle it before it starts or counteract it once it has begun. Worse for our academic world is that stereotype threat appears to lead to increased segregation in the classroom and deters threatened groups from pursuing classes or majors that fall into the stereotyped category. Steele recounts several such experiments, even someone that don’t involve academic goals (e.g., whites underperforming in tests of athleticism),and shows that not only do the threatened groups perform worse, they often perform less – answering fewer questions or avoiding certain tasks. Whether it’s women and math, blacks and intelligence, or another stereotype, the results are consistently – the ‘threatened’ group performs worse than expected (based on predetermined criteria like grades in math classes or scores on standardized math tests) when they’re reminded of the stereotype before the test. In one iteration, the test-takers are told beforehand that women tend to fare worse than men on tests of mathematical abilities in another iteration, they’re told no such thing, or something irrelevant.

The best-known proofs of stereotype threat, which Steele recounts throughout the first two thirds of the book, come from experiments where two groups are asked to take a specific test that encompasses a stereotype of one of the groups – for example, men and women are given a math test, especially one where they are told the test itself measures their math skills. The resulting book, which runs a scant 200 pages, is less academic in nature than Thinking Fast and Slow and its ilk, and thus a little less intellectually satisfying, but it’s also an easier read and I think the sort of book anyone can read regardless of their backgrounds in psychology or even in reading other books on human behavior.

Steele blends personal anecdotes – his own and those of others – with the research, mostly in lab settings, that we have to date on stereotype threat, which, again, has largely focused on demonstrating its existence and the pernicious ways in which it can affect not just performance on tests but decisions by students on what to study or even where to do so. Stanford psychology professor Claude Steele, a former provost at Columbia University and himself African-American, wrote a highly personal account of what we know about stereotype threat and its presence in and effects on higher education in the United States in Whistling Vivaldi: How Stereotypes Affect Us and What We Can Do. The effect is real, with about two decades of research testifying to its existence, although there’s still disagreement over how strong the effect is in the real world (versus structured experiments). African-American students perform worse on tests that purport to measure ‘intelligence’ for a similar reason.

For example, women fare worse on math tests than men because there’s a pervasive stereotype about women being inferior at math.

Stereotype threat refers to how common stereotypes about demographic groups can then affect how members of those groups perform in tasks that are covered by the stereotypes. In this era of increased awareness of cognitive biases and how they affect human behavior, stereotype threat seems to be lagging behind similar phenomena in its prevalence in policy discussions.

Whistling vivaldi