Speaking Truth To Test Scores

Dean of Admissions for Harvard College William R. Fitzsimmons ’67 surprised many in the academic community in September when he suggested that there may be a time in the near future when the SAT—which is currently required of all Harvard applicants—would cease to be an essential part of the application. Ditching the SAT has become trendy in recent years, as prestigious colleges like Bowdoin and Mt. Holyoke have joined middle-of-the-road schools like Pitzer and Hamilton in going SAT-optional.

Fitzsimmons’ comments came as the esteemed dean led a committee of the National Association for College Admission Counseling in examining the usefulness of admissions tests such as the SAT. According to Fitzsimmons, the SAT is no longer a good predictor of performance in college. But aside from the question of whether the SAT is an effective predictor, the SAT-optional trend is at least due in part to a favorite claim of social justice types: that the SAT discriminates against students from lower socioeconomic backgrounds. For proponents of this claim, such a statement from the venerable Fitzsimmons is an exciting turn. However, there is simply a lack of compelling evidence to suggest that the SAT caters dramatically toward students with financial means. That so many participants of this particular educational debate have become fixed on making this a class war is unfortunate, and it begins to reflect the degree to which our culture of political correctness is damaging the future of higher education.

Never mind the fact that Fitzsimmons’ proposal for an alternative would hardly alleviate socioeconomic bias. Fitzsimmons went on to say that he envisioned the SAT would be replaced by a set of five SAT II subject tests, which are considered by many to be more difficult. The subject tests rely on substance-based knowledge as opposed to skill-based knowledge, a factor that inevitably favors students from high-achieving educational backgrounds. It’s hard to imagine a student who performs badly on the SAT I because he wasn’t adequately prepared performing fantastically on five or more subject tests, which have much more to do with how well a student was taught the material than how intelligent he is.

No doubt there is a correlation between socioeconomic status and SAT scores. A survey of SAT-takers in 2003 revealed that only 14 percent of students who scored over 1420 were from families in the bottom 40 percent of income brackets. Moreover, 46 percent of those test-takers scoring over 1420 came from families in the top quintile in income level. Critics of the current SAT believe this demonstrates the degree to which test scores are determined by background and preparation. In fact, this does not actually explain the numbers. A 2005 College Board survey found that, on average, SAT tutoring raised verbal scores a mere 10 points and math scores only about 15-20 points. This is hardly a mind-blowing difference.

What these surveys may begin to indicate is something that very few are willing to entertain: there may exist a correlation between SAT scores and socioeconomic status because there is a correlation between socioeconomic status and intelligence. In his most recent book, “Real Education,” Charles Murray, a co-author of the controversial bestseller “The Bell Curve,” points out that nearly all of the most notable members of elite professions (which typically pay higher than blue-collar jobs) have IQ levels of 120 or above.

Murray is among just a handful of public intellectuals who are willing to talk openly about these statistics. Most remain fearful, because such a suggestion is so counter to our feel-good conceptions of social mobility. But there is nothing about this evidence that should make us uncomfortable. The idea that there would be a correlation between class and intelligence is completely consistent with the dream of American opportunity because it reinforces this sense that our idealized, meritocratic society is working. There is nothing shocking about the notion that people who are capable would rise to high-paying jobs. Nor is it surprising that smart, capable people would have smart, capable offspring—far more often than not, people with overlapping genes have overlapping IQ scores.

None of this is to condone leaving students from low economic backgrounds by the educational wayside. There will always be outliers, and there are surely members of lower socioeconomic classes—first-generation immigrants, for instance—who are on the academic rise. As members of educational elites, we should continue to help pave the way for these individuals by supporting college-level programs like the Harvard Financial Aid Initiative, as well as efforts for younger students, such as the successful New York-based Prep for Prep program.

However, we must also strive for honesty in our examination of these concerns. Instead of criticizing Charles Murray and contemporaries who are pushing through contrarian positions, we should embrace all the data available for us. Equalizing educational opportunities is a noble goal, but there should be nothing to fear in the truth.


Lucy M. Caldwell ’09, a crimson editorial writer, is a history and literature concentrator in Adams House. Her column appears on alternate Wednesdays.