A Two-Pronged Attack on Standardized Tests Is Needed

Article excerpt

This summer, hundreds of thousands of high-school juniors will, enroll in SAT coaching courses. They II do it because they know that a three- or four-digit number that ranges from 400 to 1600 may make every difference in where they'll go to college, and have an impact on their life chances in both the near and the distant future.

One doesn't have to take a risk to bet that African American juniors will be underrepresented in the Stanley Kaplan and other SAT coaching classes. They may also be underrepresented among test-takers. And the data shows that African Americans, regardless of grades, income level, or class rank, on average score lower than their White counterparts on SAT tests. Indeed, some have described the tests as both "racist" and "sexist" because of their inherent biases. I'd describe reliance on these tests as racist, sexist, and also stupid.

Think about it. What if White male students systematically scored hundreds of points less than African American students? Someone would be rewriting the test now! Not only would they be rewriting the test, but they'd also be involved in analysis about the flaws of the test. Instead, most people have an almost fanatic-like belief in SAT tests, even though the tests measure little. Aside from analysis from Cambridge, Mass.-based FairTest, there has been little consistent criticism of the flaws in the test.

Yet, even the proprietor and marketer of these tests -- and let's call it like it is, an economic proposition -- admits that the tests are, at best, a predictor of first-year college grades. They acknowledge that high school grade-point averages and class rank are better predictors, even given variations in the quality of high school education. They note that SATs are poor predictors of graduation rates. They are not good assessment instruments for course placement, either. They are just a multiple-choice test that skilled marketers have managed to convince universities to use in selecting students for admission.

Students' SAT scores must be at least 125 points apart before they are statistically different. Thus, when cut-off scores are used for admission, or when absolute scores are used to award scholarships, there is an inherent unfairness. Still, the National Merit Scholarships use PSAT scores to select its semifinalists. Until a court stopped them, the National Collegiate Athletic Association denied first-year students the right to compete if their scores fell short of predetermined guidelines. And many universities use SAT scores to choose candidates for gifted and talented programs.

Yet a group of students find the SATs biased against them. The ranks aren't limited to the African Americans that so many hear about. Latinos, new Asian immigrants, bilingual students, and even women do more poorly than White men do on SAT test. This hardly speaks to the "superior" intelligence of White men. Instead it suggests that some thing is wrong with the test!

Still, thousands of universities continue to use SAT trots to determine who they will admit to college, though the "verbal" section of the test does not have students write a single word, and the test hardly assesses higher-order thinking or reasoning skills. …

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.