What Does The SAT Test?

Andrew Sullivan —  Jan 16 2015 @ 6:47pm

Jeffrey Aaron Snyder gives a mixed review to Lani Guinier’s The Tyranny of Meritocracy: Democratizing Higher Education in America. In particular, Snyder finds wanting the book’s criticisms of the SAT:

Guinier, like many critics of the SAT, is dismissive of the test’s predictive power, claiming that the correlation between SAT scores and first-year college grade-point-average is “very, very slight.” In fact, most studies put the figure in the neighborhood of .45, which is a shade higher than the correlation between rates of smoking and incidences of lung cancer. It is also only a tad lower than the correlation between cumulative high school GPA and first-year college GPA. …

Guinier has been arguing for years that the SAT is a “wealth test.” Is she right? Money indisputably matters. The correlation between socioeconomic status and SAT scores is around .40. (If the SAT were nothing but a wealth test, as Guinier maintains, this figure would be 1.00.) For high school graduates from the class of 2013, students from families earning more than $200,000 a year had an average combined SAT score of 1,714 (out of 2400) compared to an average combined score of 1,326 for students from families earning less than $20,000 a year. These averages, of course, obscure the enormous variation within different income brackets—many poor students ace the test while many rich ones bomb it.

Standardized testing aside, Snyder is open to Guinier’s suggestion that “our educational system should re-orient itself around collaboration and peer learning”:

The message that we send to students through standardized testing is often perverse: we are going to assess your abilities in a vacuum, without access to the books, Internet, or peers that you will almost always have access to in the working world. In this respect, standardized testing promotes an antiquated model of teaching, learning, and knowledge. While collaboration in the workplace is rewarded, collaboration on a test is penalized as cheating. It is not just testing, however, that prizes individual achievement. The conventional classroom is organized around individual performance, with students laboring away at solitary desks. The push for more open, collaborative classrooms has always faced stiff resistance from “traditional” teachers and school administrators. If you agree with Guinier that education should be a more cooperative enterprise, then the crucial question is how to incentivize schools to embrace this cultural shift.