Reliability of SAT debated in academia

Reliability of SAT debated in academia

College Board says test is fair, others cite score
disparities

By Rachanee Srisavasdi

To most high-school students, the Scholastic Aptitude Test (SAT)
is not only just another obstacle to being admitted to college, but
is an arguably unfair measurement of how well a student will do in
college.

The SAT, taken by more than one million students annually, is
often accused of biases by both educators and students. More
specifically, dissenters claim the SAT discriminates against women
and certain minority groups, claiming that the test is geared
towards Caucasian males.

But officials at the College Board – which coordinates diversity
committees to review each new test question and has a student
advisor panel – said that the questions on the SAT are not
biased.

"How can we create a test that is biased," asked Gretchen Rigol,
director of the SAT program at the College Board, based in New York
City.

Yet looking at SAT performance averages, there is a noticeable
gap between the scoring of certain groups. Women and minority
groups score lower on the SAT, according to studies done by the
College Board itself. Men score an average of 50 points higher
overall on the SAT compared to women, and also score significantly
higher in the math section.

As for ethnic groups, Asian Americans and Caucasians score best
overall, followed by Native Americans, Latin Americans, Chicana/os,
Puerto Ricans and African Americans.

Other groups maintain these disparities show the SAT is not an
accurate predictor of a student’s scholastic ability.

"The SAT is an extremely weak predictor," said Marty Neill,
associate director of Fairtest, a performance assessment
organization based in Cambridge, Mass. "The content is more
familiar to one group of test takers than another."

But the College Board explained that differences in test scores
reflect inequity of education rather than bias in the test.

"If you look at average scores, there are differences," Rigol
admitted. "But, just because you have differences, does not mean
the SAT is biased."

Admissions officials agreed that the SAT performance depends on
a student’s educational opportunities.

"What the test reflects are differences in a student’s
educational opportunities," said Rae Lee Siporin, UCLA’s
undergraduate director of admissions. "If you go to a high school
with tough curriculum, you’re more prepared for the test as opposed
to other students."

Yet some analysts said that depending on how one defines bias,
the SAT might not be completely reliable.

"If you measure the SAT as an accurate predictor of future
grades in college, then the SAT is biased," said Lori Shepard, a
professor at University of Colorado who has researched testing and
assessment.

On average, while women score lower on the SAT, they get better
grades in college as opposed to men, who score well on the SAT but
tend to have lower grades in college, Shepard said.

While her studies show that SAT scores are not an accurate
predictor of college performance, the tests are a crucial criterion
in college admissions. UCLA admission officials agreed that SAT
scores play a key factor in deciding which applicants to admit.

"The SAT is the only national test to evaluate student
performance," Siporin said. "It’s an opportunity beyond the GPA to
see how a student performs."

Yet some students said that they do not like the weight given to
SAT scores by colleges.

"I don’t think SAT measures intelligence," said Tracy Tuyen, a
second-year development studies student. "It measures only how you
do on tests."

Other students thought the SAT had little to do with actual
intelligence.

"The SAT is a game," said Mark Trang, a second-year political
science student. "There are a lot of shortcuts and test-taking
strategies a student can use to get around the test."

SAT preparation classes, such as Princeton Review and Kaplan,
are another issue of equal opportunity. Some said that students who
take costly SAT preparation classes have an advantage over other
test takers.

"The fact is that the SAT is coachable, which creates a bias,"
Neill said.

But the College Board argued that a student can do just as well
by studying independently.

"The advantage is being prepared, not taking prep classes,"
Rigol said. "You can do it yourself, if you’re responsible."

Responding to educators who said the SAT needed change, the
Educational Testing Service (ETS) revised the SAT to "recognize the
increasing diversity of students " and accommodate curricula taught
in secondary school.

Changes in the new SAT – which was introduced in spring 1994 –
increased time intervals from 60 to75 minutes per section and added
more questions that focused on critical reading of passages. In
addition, a new math grid question requires test takers to produce
their own answers .

Antonym questions were also dropped and students could use
calculators for the math section.

Most high school students like the new revisions of the SAT,
especially the option of using questions and the grid questions,
according to the College Board.Comments to
[email protected]