Via a Wall Street Journal story of August 31, the Educational Testing Service (ETS) of Princeton, New Jersey, the public an advance peek at a research initiative called Strivers, which could give underprivileged takers of the Scholastic Aptitude Test (SAT) credit for exceeding expectations.
The expected SAT score for an individual student would be determined by considering more than a dozen personal factors, such as family income, mother’s employment status, language spoken, quality of school attended, race and ethnicity, even the kinds of appliances in the home. The more disadvantages conferred by these factors, the lower the student’s expected SAT score. By the same token, the fewer disadvantages, the higher the expected score.
If a student scored 200 points higher than expected on the combined math_verbal SAT, he or she would be deemed a Striver by the proposed ETS measure.
And then what? College admissions offices that used the ETS gauge to make their own calculations next would determine what advantage, if any, to give Strivers in their pool. It would be up to admissions offices to decide whether to use race as one of the factors. Factoring in race for those with SAT scores of 1000 to 1200 would increase the proportion of black and Hispanic Strivers from 5.6 percent to 8.4 percent, but it would reduce Asian Strivers from 3.9 percent to 3.3 percent.
After the Journal story unleashed a torrent of comment and other media coverage, the ETS issued a “clarification” over the PR Newswire, contending that news coverage had created a misimpression that the initiative would alter the SAT scores of Strivers. To the contrary, said the ETS, researchers are merely studying the effect of considering background to “provide a richer context for candidates’ scores,” and a final report won’t even be released for another two months.
Clearly, though, the research attempts to quantify expected performance, and that raises a number of questions. What about a student who lives in a comfortable home but has struggled academically? Can that student never be a Striver, even after working hard to master, say, algebra?
What about the starting premise that society has low expectations for low_income minority students? Dartmouth admissions dean Karl Furstenberg thought that could have a negative effect on some students taking the SAT.
“When we make admissions decisions, we already take into account a student’s background when evaluating their scores,” said Furstenberg. “I don’t like this formalizing of it__calling people ‘strivers’ just because they do 200 points better than expected.”
Some other admissions officers, though, gave the ETS higher marks. Joyce Smith, executive director of the Alexandria_based National Association of College Admissions Officers, agreed that admissions people have been doing something similar all along. They dub it “adversity overcome.” But that was, she said, “subjective,” and “I’m fascinated that ETS has found a way to quantify those things.”
The idea of Strivers comes after a summer of heated debate on Capitol Hill over a draft policy from the Education Department’s Office for Civil Rights that suggests schools and colleges should go to alternative assessments if standardized tests have a “disparate impact” on minority and female test_takers. The ETS research also comes amid controversies over universities’ attempts to preserve preferential treatment of minority applicants despite state initiatives and court decisions rejecting affirmative action.
Referring to the Strivers project, Terrence Pell, senior counsel at the Center for Individual Rights, asserted, “This is race_norming [of SAT scores] in another guise. I think there will be legal difficulties with that.” The Center has led court battles against race and sex quotas in higher education.
“Race_norming” was a practice used by federal and state employment services in the 1980s, whereby scores on aptitude tests were weighted to give black and Hispanic job seekers a boost. Congress outlawed the practice in the 1991 Civil Rights Act.
Robert Holland is a senior fellow at the Lexington Institute, a public_policy think tank in Arlington, Virginia.