Psychology Today ponders this question in their online magazine:
The SAT has long been a three-letter word for aspiring college students—an abbreviation that stands for fear. It has become the single most important arbiter of university admissions, the magic number that seems to predict the future for nervous teens. But the tide may be turning, at least among elite private schools. The latest nail in the coffin: Sarah Lawrence College, a prestigious liberal arts school in Bronxville, New York, recently announced that beginning with the high school graduating class of 2005, applicants will no longer be required to submit standardized test scores.
The decision was intended “to reflect our belief that standardized testing is not effective in evaluating a student’s ability to succeed in a writing-based curriculum such as ours,” the college’s dean of admissions said in a press release. The statement also fretted over the growing inequity between test-takers who can afford preparation courses (which can cost up to $900) and those who cannot.
Not a bad beginning, but I already have several comments:
(1) Is the SAT score really the "single most important" factor in the admissions decision-making process? Almost certainly not, especially for lower-tier schools, and not in this age of affirmative action and "comprehensive review."
(2) Is the decision by Sarah Lawrence really the sign of a trend against the SAT? After all, the college isn't opposing the SAT on principle; it's perfectly legitimate for a college to base grades on writing skills and then to decide that the SAT isn't a good predictor of those grades, for that college. There's no reason to suggest that this "trend" will catch on with colleges that place emphasis on other types of performance.
(3) Tuition at Sarah Lawrence College is over $28,000 a year. And we're supposed to believe that they're really worried about applicants who can't scratch up a measly grand for test prep? That sum isn't even necessary - a prospective student can prepare for the SAT using 10 practice exams for $19.95. Come on. Sarah Lawrence is more than entitled to decide that the SAT is not right for their students, but not on the grounds of expense.
Anyway, let us continue...
The SAT is the nation’s oldest and most widely used college admissions test. It is also big business: The College Board earns millions each year in revenue from the exams.
And your point is? Is the College Board supposed to give the test for free?
But its critics have gained momentum in recent years. They say that the SAT is a poor predictor of women’s, non-native English speakers and older students’ academic performance in college. Whites outscore African Americans on average by 206 points.
"They say"? Come on, PT, you have to do better than that. "They" can say anything; whether they back it up with data is a different story. What's more, the utility of the SAT can vary from school to school, and it's inaccurate to claim that, for example, the SAT does not predict well for women across the board. The mention of White scores vs. African American scores is a non-sequitur here, regardless of its accuracy. The score gap is related to differential impact, but not to bias; nor is the gap any indication of whether the scores for both groups are differentially useful.
When the article does get around to quoting someone, that person is, unsurprisingly, a spokesman for an anti-testing organization. The National Center for Fair & Open Testing hastens to mention that "700 colleges and universities" no longer require the SAT, but the reporter should have wondered how many of those were recent changes, and should have noted that those 700 comprise less than 17% of the 4182 private and public two- or four-year institutions in the US.
The only other person quoted is a Yale psychology professors who thinks the SAT isn't adequate for predicting college performance. Neither testing critic provides studies to back up his claims. And the conclusion is a load of anti-wealth and multiple intelligence blather:
Sternberg says that the SAT tests well for memory and analytical skills. “Clearly, these are important,” he says. “But in life, and in college, you need more than analytical skills.” The test is not so good at gauging creative skills and practical skills, which means that some kids may not be fairly ranked by the exam. “A kid from a challenging environment might have better-developed creative or practical skills, whereas a kid who grew up in Scarsdale [may have] had the luxury of developing analytical skills,” says Sternberg. His recommendation: Admissions officers should learn how to recognize a diverse set of skills or risk rejecting talented applicants.
Does Sternberg really believe that admissions officers don't already do this? No university in the US relies solely on SAT score for admissions, and College Board policy is that the SAT should not be used as the sole method of selecting students. What's more, the reference to "Scarsdale" is ridiculous. Depending on the university and the program, "creative skills" may well count for nothing when compared to the need for analytical skills. Sternberg might think that it is a "luxury" for a student to have had the opportunity to develop analytically (if true, that's an indictment of the K-12 public school system, not a criticism of the tests), but it's just plain silly to conclude from this statement alone that analytical skills should be downweighted in the college admission process.
C'mon, Psychology Today. You didn't even make a half-hearted stab at even-handedness. One measly quote from someone at the College Board - or from any psychometrician - would have gone far towards making this article something other than a one-sided essay about the potential unfairness of standardized tests. For example, I could have pointed out that the artificial restriction of range of the SAT scores of admitted students could be depressing the SAT-college grade correlation. Also, it's a crime to write an SAT article and not quote expert Wayne Camara, who has published research supporting, among other things, the usefulness of the SAT in predicting college grades for women. He also notes that the claims that huge numbers of colleges are dropping the tests may be greatly inflated, and says that between 92% and 96% percent of four-year institutions require an admissions test.
There ARE valid reasons for a university not to require the SAT, so I don't have a problem with the idea of pondering the utility of the SAT. It's just a shame to see an article waste the reader's time with the unsupported and one-sided arguments made here. What's more, I'm not even sure what conclusion the reader is supposed to draw from this article.
For example, if the SAT suddenly went out of favor for all universities, the College Board would simply develop another test more to the liking of the administrators (in fact, that's how the new SAT revisions came about). Five years from now, the College Board might pop up with a new writing exam that better predicts student performance at colleges like Sarah Lawrence. If the writer was hoping this article would make readers root for the bankruptcy of the College Board, she's got it all wrong.
Posted by kswygert at January 7, 2004 09:58 AM