SAT I has little effect on admissions, study says

By Noah Grand
Daily Bruin Reporter

A University of California study released today found that
omitting the SAT I from the UC admissions criteria would not solve
problems of socioeconomic and racial inequities.

The study reports that 22.2 percent of the variance between
college freshman’s GPA is explained by their high school GPA
and SAT II scores combined. Only 0.1 percent more of the difference
is explained when SAT I scores are also considered.

“These findings suggest that eliminating the SAT I in
favor of the SAT II … would have little effect on the rates of
eligibility and admission among students from different
racial/ethnic groups,” the study states.

The study also suggested that any correlation between SAT I
scores and college freshman GPA is determined by that
student’s socioeconomic background. The SAT II, however, is
less dependent on this background.

UC researchers Saul Geiser and Roger Studley, who conducted the
study, began working on the report in March, after UC President
Richard Atkinson proposed to the UC Academic Senate in February to
eliminate the SAT I as a requirement for UC admissions. The SAT I
has been criticized by some as racially and socioeconomically
biased.

UC spokesman Brad Hayward said race was never the reason
Atkinson proposed to oust the SAT I as a requirement, though many
proponents have seen it as such.

The UC is in a unique position to compare the usefulness of the
SAT I and II to each other because it is the only major university
system to use both tests. The UC requires SAT I scores and three
SAT II subject test scores.

According to the study, the SAT II is a preferred subject test
because it reflects high school curriculum more and is based less
on socioeconomic factors.

“Performance on both the SAT I and SAT II are correlated
with socioeconomic background,” Geiser said. “But the
SAT II is a fairer test.”

He said students from low-income backgrounds ““ often
underrepresented minorities ““ usually receive a lower quality
K-12 education and cannot afford SAT I preparation courses, leading
to lower test scores. Some also criticize the SAT I for testing
things that aren’t in the curriculum.

Geiser and Studley tracked close to 78,000 students entering the
UC between 1996 and 1999 and compared their high school GPA, SAT I
and SAT II scores to their GPA in their freshman year of college,
looking for a correlation between the two.

Max Espinosa, a higher education policy analyst for the state
assembly budget committee, said he is not surprised by the
findings.

“A lot of these standardized tests are inherently
discriminatory,” Espinosa said. “If you substitute the
SAT I with some other standardized test, it’s probably likely
you would get very little difference.”

Geiser started researching whether the SAT was an effective test
two years ago for the Academic Senate’s Board on Admissions
and Relations with Schools. This earlier research indicated that
achievement tests of present knowledge, such as the SAT II subject
tests, are more effective than aptitude tests such as the SAT
I.

Because of this, BOARS changed the UC eligibility index, the
formula used to determine minimum UC eligibility requirements, for
fall 2001 admissions.

For the first time, SAT II scores were considered in UC
eligibility and given twice as much weight as the SAT I.
Previously, students were required to take the SAT II to be
admitted into the UC, but there was no minimum requirement on these
scores.

But the ethnic breakdown of UC eligibility changed very little
despite placing more weight on the SAT II. The difference in test
scores among students of different ethnic groups is also “too
small to have any substantial effect on the demographic make-up of
the UC admissions pool,” according to the report.

The study also shows that the third SAT II subject test, which
gives students an option of 19 different subjects to be tested on,
is still a more useful predictor than either the SAT I verbal or
math test.

There are concerns that students will choose a foreign language
they learned at home for the third subject test instead of a
subject they studied in school.

Some say these students have an advantage in gaining UC
admission because they are taking the third subject test in a
native language.

BOARS is considering proposals to change standardized testing in
the admissions process, including developing a new standardized
test, using five SAT II tests or limiting the use of SAT language
tests in admissions.

Studley said changing tests was only one step in solving the
inequities in the admissions system.

“The real way to improve balance in education among people
from different socioeconomic backgrounds is to improve the balance
at local schools,” he said.

“Shifting the focus on to achievement tests is a step in
the right direction.”