In defense of the SAT

To standardize or not to standardize? (photo by James Sarmiento)

Josh Anish is the Senior Editor at Knewton, where he helps students with their SAT prep.

–

The SAT was introduced seventy-five years ago to measure aptitude. Whereas most tests look backward at acquired knowledge, the SAT was designed to predict a high school student’s future academic prowess. The goal was to open doors for promising students based on merit, not on social status or past educational opportunities. This basic premise has been under fire ever since.

The latest challenge has come in the past few years, as several more schools have joined the ranks of those who no longer demand SAT or ACT scores as part of the admissions process. Wake Forest University became the first top-30 national university to make the SAT optional for applicants. Some schools claim that not requiring test scores levels the playing field, giving more opportunities to people of varying socioeconomic classes. This assertion is supported by a research study conducted by Professor Thomas J. Espenshade and Chang Young Chung at Princeton University. Their study concluded that dropping standardized test scores as an admissions requirement leads to increased acceptance of minority and disadvantaged students. In doing away with the SAT are we shooting the messenger?

We should applaud institutions for their desire to have more diverse student bodies. But there are drawbacks to eliminating the SAT, both for students and for the educational system.

Yes, the SAT has its flaws and limitations. But no measurement of aptitude is perfect. Without the SAT, colleges need to place more emphasis on high school grades, essays, extracurricular activities, and interviews. A student’s grades are subject to the vagaries of different teachers’ standards, which rely on different baselines and expectations. It is exceptionally difficult to adjust for grade inflation. Personal essays, recommendations, and interviews are even more subjective and variable. And how can admissions officers compare extracurricular activities? How can they distinguish genuine commitment and interests from meaningless resume filler?

Standardized tests can never be completely free from error, but they can offer more objectivity than can other indicators, since all test-takers face the same challenge. In the case of the SAT, all test questions are routinely pre-evaluated for potential measurement bias against members of different minority groups, and any questions found wanting are eliminated before they become operational. The SAT continues to be the nation’s most widely-taken standardized college admissions test. Last year, the number of SAT takers rose to more than 1.5 million, an 8% increase from 2003 and a 29.5% increase from 1998.

Critics point out that several studies show that the correlation between SAT scores and first-year college grades is not overwhelming; SAT scores only account for ten to twenty percent of the variation in first-year GPA. However, the true correlation is stronger than these studies suggest. Most colleges accept students from a fairly narrow section of the SAT spectrum. The scores of students at elite schools are considerably higher, on average, than those of students at community colleges, even though both sets of students are likely to have similar college grade distributions at their respective schools.

If both sets of students were admitted only to elite schools or only to community colleges, there would be a much stronger correlation between their SATs and college grades. It’s like comparing the average height of ballerinas to that of NBA players. Therefore, schools that accept students with a wide range of SAT scores have even tighter correlations between scores and first-year grades. In addition, myriad studies have shown that the combination of a student’s SAT score and high school GPA is the single best academic predictor of first-year college performance.

Another criticism of the SAT is that it primarily tests how well a student prepares for the SAT. This seems to undercut the SAT’s promise to test scholastic aptitude. We would like to be able to distinguish people who achieve based on innate ability from those who simply work harder or receive better instruction. But is it possible to separate ability from effort? Studying for a test is a good model of education generally. Navigating the SAT is a particular skill, and like any other skill, it takes practice. Taking sample tests or enrolling in an SAT prep course will maximize any prospective test taker’s chances of success.

No single standardized test by itself should be considered a complete representation of a student’s academic aptitude, but neither should any other single measurement. Admissions officers should consider the most diverse set of data available–including grades, essays, and standardized tests–to create the most complete picture of each applicant.

The SAT is not perfect, but neither is any single piece of an applicant’s profile. Eliminating any one part of the application–including the important and standardizing SAT–would result in worse measurement, not better, and it would introduce bias and arbitrariness into the admissions process instead of reducing it.

Read the full article: In defense of the SAT

SPONSORS

POPULAR ARTICLES

Related Articles