Search News


Browse Archives

News

SAT Failures Renew Suspicions

March 16, 2006

Share This Story

FREE Daily News Alerts

Advertisement

As a result of thousands of miscalculated scanning errors on SATs taken by students in October, traditional critics of the exams are seizing the opportunity to blast standardized testing. College Board officials are taking the blame for the mistakes, but say that race- and gender-based arguments against the test are equally miscalculated.

“Those who are gravely suspicious of standardized testing now have another route to attack the reliability of tests like the SAT,” says William C. Hiss, vice president for external affairs at Bates College, in Maine. “People have long held the accuracy and scoring of the SAT as solid. The questions of the past had always centered on predictability, not accuracy.”

Hiss’s institution isn’t an SAT-hater. In fact, about three-quarters of its students submit the scores each year, even though Bates hasn’t required them since 1984. Bates was one of the early competitive admissions institutions to drop the SAT as a requirement.

“We might want to give the College Board a little slack,” says Hiss. “But at the same time, it’s hard not to recognize the accelerating numbers of institutions that don’t require them anymore.”

Hamilton College just announced that an experiment making the SAT optional had been a success -- and would now be the college's policy for good. Hamilton found that the policy helped attract more minority applicants -- and that those students who didn't submit SAT scores fared slightly better at the college than those who did.

According to FairTest, a standardized testing watchdog group, more than 700 institutions do not require the SAT or ACT to make admissions decisions. Robert A. Schaeffer, director of the organization, argues that administrators have chosen to deemphasize the tests because “the tests don’t work.” He asserts that many individuals and researchers have found racial and gender biases in the questions of the SAT in particular. 

“This College Board scenario is the one that may break the camel’s back,” he says. “The scope of this scoring error is unprecedented to my knowledge.”  College administrators nationwide learned only last week that the scores of 4,000 students who took the test last October were miscalculated, even though the College Board had learned of the situation in December. This week, the company announced that another 1,600 tests had been were omitted from examination and must be reviewed. Newspapers have been filled with stories of confused students and angry college admissions officers. 

Schaeffer predicts that the coming years will see an uptick in the number of institutions that will eliminate standardized testing requirements. His organization, which has faced financial challenges in recent years, has received an outpouring of public donations since the College Board's latest problems surfaced, he says.    

Phyllis Rosser, director of the Equality in Testing Project, has argued for decades that gender and racial biases exist within standardized testing. “Certain questions favor one sex over the other,” she says. “The same with race, and the College Board knows this.” 

“Math scores are really where women fare worse, but they get better math grades in high school,” says Rosser. “If they’re going to be used, these tests have to be made more predictive.”

Rosser argues that the scoring miscalculations should make people realize that standardized testing is not infallible. “We don’t know that this hasn’t happened in years past,” she says. “I think College Board was going along thinking everything was fine.”

Brian O’Reilly, executive director of SAT information services at the College Board, says that both the College Board and Pearson Educational Management, the vendor that grades the SAT, are “equally to blame” for this year’s miscalculations. “The worst thing about all this is the timing for students,” he says. “It’s already a very stressful time for them, and all I can say is that we alerted them as quickly as we could.”    

But O’Reilly insists that the renewed attacks from people like Schaeffer and Rosser are misguided. “Critics who say the test is biased against minorities should look at the results of minority students in college,” he says. “The SAT actually overpredicts how well they will do when they get to college. It’s not biased simply because scores tend to be lower for one group than another.”

Rather, says O’Reilly, the real problem probably lies in the teaching of minority and female students in elementary and secondary education.

Still, increasing numbers of administrators are making research-based arguments for a shift away from standardized testing. “High-stakes standardized tests such as the SAT have assumed a central role in the admissions process disproportionate to their value,” Joanne V. Creighton, president of Mount Holyoke College, in Massachusetts, said in a Los Angeles Times editorial this week. “This test falls far short of predicting academic or career potential or a host of important aptitudes, such as curiosity, motivation, persistence, leadership, creativity, civic engagement and social conscience.”

In 2001, Mount Holyoke administrators chose to make the SAT optional for admission.

“The SAT might have made sense when it was developed in the 1920s, when higher education was an elitist proposition and the college admission pipeline led a relatively homogeneous population of young adults into a similarly uni-dimensional set of colleges and universities,” Creighton wrote. “But U.S. secondary education today is a multilingual, multiethnic, socioeconomically diverse enterprise, and so too are the 3,000-odd colleges and universities to which high school students aspire.”

Creighton said that her institution has been studying the effects of making the SAT optional. To date, she indicated that there has been no meaningful difference in academic performance between students who did not submit scores and those who did.

Hiss says Bates College has come to similar conclusions based on research. “Optional testing has profoundly changed Bates,” he says. “We doubled our applicant pool. We are more diverse, come from more states and countries, and from wider socioeconomic backgrounds. And I think more colleges are going to follow our lead.”

See all postings »
Advertisement
Advertisement

Matching Jobs

Comments on SAT Failures Renew Suspicions

  • SAT Testing
  • Posted by Mike on March 16, 2006 at 12:31pm EST
  • I believe that the debate taking place over the SAT as a result of the recent scoring problems is healthy, and that points from both sides of the argument have merit. But what I wonder is what will happen to the admission process if we eliminate the use of standardized test scores. It’s great to read about the schools that are dropping the SAT (where the heck is the ACT in all of this) and will be approaching their admissions processes with a more holistic approach; using personal essays, interviews, teacher recommendations, etc. to make their admission decisions. But my question is who is going to work with all the students to put together an application packet that paints a clear and accurate picture of the student if we no longer use test scores? We don’t have enough high school counselors as it is to take care of the thousands of students who are missing out on important college information and help with the current process. If we eliminate test scores and ask for more indicators of student performance to determine admissibility we have suddenly created a new process that will be incredibility demanding to implement.

    Maybe what we need to bring to the discussion is some common sense that says we keep the test scores in the admission process but realize it is only ONE indicator of student performance and it should be use that way. The trouble we get into is in using the test scores as cut scores to narrow down an application pool thereby clearly eliminating low-income and minority students who typically score lower. Now that is bias. By the way the only group I hear making the argument to eliminate the use of test scores are the private elite schools. Where are the large state universities with those 35,000 applications to make decisions on in this discussions?

  • Test Optional College Admissions
  • Posted by Robert Schaeffer , Public Education Director at FairTest: National Center for Fair & Open Testing on March 16, 2006 at 1:50pm EST
  • Test optional colleges generally require neither the SAT-I nor the ACT. Check out the updated list posted at: http://www.fairtest.org/optinit.htm

    The 731 accredited, four-year degree granting schools on this list evaluate applicants by looking at their entire portfolios, including high school grades, class rank, rigor of courses taken, extracurricular intersts, community service, leadership, family background, etc. They know that these factors provide a richer portrait of a candidate than how well he/she scored one Saturday morning. And, high school record is much less susceptible to the impact of high-priced coaching

    The reason why most of the growing number of schools that have dropped admissions test score requirements are selective liberal arts colleges is that these institutions are governed by administrations and faculties motivated by educational values consistent with their unique missions. They have come to recognize that requiring test scores can be inconsistent with their other goals. Admissions policies at public universities, on the other hand, are often regulated by politicians who have embraced the false notion that "test scores equal merit" or simply see ever-higher test score requirements as a cheap way to give the appearance of quality.

  • Posted by Cindy on March 16, 2006 at 2:40pm EST
  • To Mike,
    The answer to your question is the University of California (the SAT’s biggest customer) in 2002 threatened to drop using the test which prompted the “New” SAT last March.

  • Admission Counselors
  • Posted by David Robertson , Professor at SUNY on March 16, 2006 at 6:10pm EST
  • Now they are caught red handed – the whole world now knows SAT tests are corrupted.

    What would admission counselors do? I assume they will rely on those corrupted scores. They are used to finagle the composition of the student body and exclude African Americans.

    College Board has embarked on rally round the flag even with corrupted test. CB lame excuses ‘moist’ caused the snafu; a better one would be ‘the dog ate my homework’

    If you believe in moist excuse, I have Brooklyn Bridge for sale.

  • Standardized Tests
  • Posted by Dr. F. Gump on March 16, 2006 at 8:20pm EST
  • Oh good grief.

    Standardized tests along with class rank (subject rating of course rigor), community service, etc. etc. still can be used to predict 84 - 86% of the variance in student success in college.

    The colleges that are doing away with standardized testing for admissions are trying to fill seats in the freshman class; period.

    Looking on the bright side, instead of just accepting the cream of the high school crop and polishing the results for four years, colleges are now finding they need to open the doors to all. (this is bright news for parents, but professors are gloomily finding that they now need to do more teaching, learning activities, one on one mentoring, etc. and less research)

  • Correcting the False Statements of "Dr. F. Gump"
  • Posted by Bob Schaeffer , Public Education Director at FairTest: National Center for Fair & Open Testing on March 17, 2006 at 1:35pm EST
  • The claims about the SAT and college admissions of Dr. Gump are as fictional as the experiences of the movie character whose name he happens to share.

    Even the College Board does not claim a correlation in excess of .6 or so in predicting first year undergraduate grades from the optimal mix of SAT scores and high school performance. That would mean that the maximum variance predicted is in the range of 35% - 40%.

    The assertion that the growing number of test-optional institutions are "just trying to fill seats" is contradicted by their acceptance practices. Bowdoin, for example, rejects more than three out of every four applicants. Many others are rated "most selective" or "very selective."

    As this week's SAT misscoring fiasco demonstrates, blind faith in the value of testing needs to be tempered with hard data from real world experience.

  • Part of the Picture...
  • Posted by Carol , Academic Counselor at University of Iowa on March 17, 2006 at 3:30pm EST
  • A test score represents information aboout a student on a specific day and as such should have always been considered only as a fraction of the information about a student. Other factors should have always been considered and now that testing has come under fire, we're finally jumping on a wagon that we should have always been riding and driving. I'm just glad some folks have caught up with testing reality, not to mention good ol' common sense.

  • Standarised exams here to stay for obvious reasons
  • Posted by Thomas Simmons , Dr. on February 18, 2007 at 7:41pm EST
  • Here we are nearly a year after this was posted. The Korean debacle over using old exams has brought the SAT yet another black eye, but students all over the USA are still gearing up, states are paying for the exams, students are taking the PSAT and the SAT and worrying about their scores. As long as there is a great deal of variation in the abilities of the secondary ed graduates and the quality of the education delivered in secondary education schools, and as long as post-secondary institutions desire to teach the teachable and maintain, at the very least, a veneer of quality in their own education, then there will be an attempt at establishing minimum criteria below which the applicants will be rejected. To do this with any really accuracy and fairness and meet the criteria the USA elucidates in legislated mandates for human rights will require an attempt at reaching an elusive goal: Screening out the unwilling and the incapable so that the efforts of post-secondary institutions to teach and prepare students for life after tertiary education are truly effective. Those who attempt to reach this goal are only too aware of how difficult and time consuming this can be and as a result, expensive. For these reasons, any evaluative measure that purports to support this goal, and save time and money will continue to have great appeal. Whatever happens to the SAT, whether it falls from favour or weathers the problem, these measures will continue to be employed. If the SAT is eventually eliminated as a required instrument, another will take its place in time as the burden of selection takes a toll on time and resources.