In particular, the Scholastic Assessment Test has been the focal point of recent debates on the fairness of traditional admissions exams. SAT opponents say that the test scores tend to reflect the income and education levels of parents, that males score better than females in certain subjects, and that white students consistently score better than Latinos and blacks.
The University of California systemcommited to attracting, enrolling, and graduating a student body as diverse as the population of Californiais considering a proposal to drop the use of SAT scores from the admissions criteria following the adoption of SP-1, a UC Regents resolution that eliminates the consideration of race and gender in UC's admission process.
"It's amazing to me that a test that only takes three hours could have such an effect on peoples' lives," says Richard Duran, one of the University of California, Santa Barbara Graduate School of Education faculty members who provided background on the debate.
Duran points out that it is difficult for some students to afford the fees to take the SATs more than once. The SAT I: Reasoning Test alone, which includes a basic registration and reporting fee to four colleges and scholarship programs, costs $22.50. Additional costs include those for writing test, subject tests, language tests with listening, and further registration and score reporting fees, which could potentially cost a student over $100.
Charged with recommending ways to achieve diversity within the context of SP-1, the UC Latino Eligibility Task Force questioned the use of SATs as an admission requirement in their fifth report, published in July 1997, because Latino students consistently perform poorly on SATs, yet their graduation and retention rates vary little from the UC average.
The UC task force, which Duran served on along with colleague Reynaldo Macias, reported that even though Latino families place great value on higher education and Latino students represent a large and growing population in California, the number of Latino students applying to UC schools is decreasing.
"I believe there are some uses for which standardized testing can be positive. However, I also believe that we have relied on this kind of testing far too heavily for high stakes decisions, particularly at the university level," says Macias, a former director of the UC Linguistic Minority Research Institute.
Rebecca Zwick, who came to UCSB in 1996 after 12 years as a researcher in the Statistical and Psychometric Research Division at the Educational Testing Service, says that while she supports the idea of re-evaluating and broadening admissions criteria, research shows that the SAT does contribute to the prediction of first-year grade-point averages in college.
"One legitimate criticism of this kind of research," says Zwick, "is that this is not the only possible indicator of college success. It's the most widely available one, and that's why it gets used the most."
"The academic world would all tend to agree that standardized testing is a useful tool if used correctly. It is how policymakers and others use the tool that should cause concern," says Jon Snyder, director of the Teacher Education Program at UCSB. "It does not matter what test they use, any single, time-bound measure is a problem. Wise decisions require multiple sources of evidence collected over time, which is what UC has historically done."
A new UC analysis of the proposal to disregard SAT scores has found that it may not yield the desired results. Without SAT scores, there would be a heavier reliance on high school grade point-averages, and in order to keep UC's eligibility pool at the same target level of the top 12.5 percent of California's high school graduates, the minimum GPA for eligibility would have to be raised to 3.65 according to the study. Under this projection, African American eligibility rate would decline by 18 percent, white eligibility rate would increase by 17 percent, Latino eligibility rate would increase by 5 percent, and Asian American eligibility rate would drop by about 1 percent.
"Instead of focusing on what standards we could eliminate, I'd like to see us try to define and measure the characteristics and accomplishments of students who are talented but don't 'test well' because of a lack of educational opportunity or for other reasons," adds Zwick. "I certainly believe that there are such students, and to attract them we really need to think more specifically about the desirable qualities they have which may not be reflected in their SAT scores."
Duran, who was also a former research scientist at ETS for seven years, feels there needs to be more emphasis on educational opportunities and quality of education prior to college.
How necessary is testing? Snyder cites a Progressive Education Association study conducted in the 1930s known as "The Eight Year Study." Operating on the assumption that students were taught only what they would be tested upon, the study sought to find methods that would support students' critical thinking skills, social sensitivity, and invigorate genuine interest as well as basic academic achievement. The study, conducted from 1933-41, evaluated 30 high schools whose graduates received college dispensation from standardized exams and prescribed courses. The researchers found that the graduates from these high schools, who were assessed through four years of high school and four years of college, outperformed their peers from more traditional high schools on every measure.
"At UCSB, there are many faculty members who are concerned about student diversity and act on it in terms of recruiting underrepresented students and involving those students in research programs," says Duran.
"Although we have a big job to do as far as access to the university for underrepresented students, we have programs here that facilitate faculty and staff collaboration in improving diversity at UCSB and other universities," he concluded