A new paper has rekindled one of the most controversial questions in the long history of the nation’s most famous test: Is the SAT racially biased?
In 2003, Roy O. Freedle, a retired senior research psychologist at the Educational Testing Service, took up the question in an article published in the Harvard Educational Review. His conclusion was that black students often do better than white students of similar ability on difficult SAT questions, but that they do worse than their white counterparts on easy items. He suggested that easy questions use a common vocabulary, making them more open to interpretation based on a test taker’s cultural background.
Mr. Freedle’s conclusions were rejected by the test’s backers, the College Board and the Educational Testing Service, which have long maintained that the test is not biased.
Now, two researchers who replicated Mr. Freedle’s methodology have published their findings in the latest issue of the Harvard Educational Review, as Jay Mathews of The Washington Post reported on Thursday. The researchers conclude that Mr. Freedle did detect a flaw in the SAT.
Some verbal questions, the researchers wrote, do function differently for the black and white subgroups: “The confirmation of unfair test results throws into question the validity of the test and, consequently, all decisions based on its results. All admissions decisions based exclusively or predominantly on SAT performance … appear biased against the African-American minority group and could be exposed to legal challenge.”
College Board officials told the Post that the new paper was flawed, and that ETS plans to publish a criticism of its findings. In the ongoing testing debate, some questions don’t yield answers — only more questions.