May 3, 2006

New SAT Mystery

The College Board is trying to figure out why many colleges are experiencing declines in their average SAT scores this year.

Looking for a job?

See all 211 new postings

Browse all job listings:

Faculty: 3,532
Administrative: 1,671
Executive: 168
FEATURED EMPLOYERS

In a memo to admissions officers, the board said that it believed that there was a four-to-five point decline, on average, comparing scores this year to last, excluding the new writing test. A “small additional decline” may be evident when all scores for the year are compiled. Based on the size of the decline, colleges should feel that it is “valid” to compare students’ scores on this year’s exam to those from previous years, the memo said.

But many colleges are reporting larger declines — in the range of 10-20 points. And several admissions officers said that they were concerned because other measures of applicant quality — class rank or courses students take in high school — were either holding steady or improving. Typically, admissions officers said, such measures and SAT average scores would rise and fall together. And the College Board had assured colleges repeatedly prior to the changes in the SAT over the last year that score comparability would not be hurt.

“It’s certainly curious and unusual,” said Stephen Farmer, director of undergraduate admissions at the University of North Carolina at Chapel Hill, where the average SAT score is down 12 points this year — while all other measures of applicant quality are up. Farmer said that officials of the College Board have been “very diligent” in working to try to figure out why scores dropped and what it means.

Similar declines have been reported systemwide at the University of California, which has not seen declines in measures of quality aside from SAT scores.

Privately, some admissions officials were less charitable than Farmer is, noting that they were hearing from colleague after colleague about larger drops than the College Board was reporting. For the College Board, any new concern about the SAT couldn’t come at a worse time. The board has been facing criticism, ridicule and at least one lawsuit for scoring errors on the October 2005 SAT, and for a series of announcements in which the acknowledged extent of the scoring errors grew.

On Tuesday, Gaston Caperton, president of the board, was defending its procedures at a New York State Senate hearing called over concerns about the accuracy of scores. Caperton’s testimony did not mention the latest concerns.

In its memo to admissions officers, the College Board said that the declines may be caused by “changing test taker patterns, including a trend of some students taking the test less frequently.” The board said that “we believe that this decrease in repeat test-taking may account for some of the average score decline” because re-tests commonly increase a score by 30 points on reading and math combined.

Farmer of Chapel Hill said that he didn’t think applicants would be hurt because they are competing against one another.

But other admissions officials aren’t sure. Bob Voss, dean of admission at La Salle University, in Philadelphia, said he is concerned. His institution saw a 10 point drop. In talking to his colleagues at other campuses, he said, “I haven’t found anyone who thinks it’s true” that the score decline has been as small as the College Board has said. Voss said that his applicants this year are notably improved in class rank.

One person’s SAT score and class rank may not correlate as expected, he said. “But you can look at 5,000 of them and it means something,” he said. “I’m seeing scores that are just amazing to me.”

And Voss said that many colleges do make decisions based on SAT scores. At his institution, for example, some students who are admitted with relatively low SAT scores are encouraged to enroll in enrichment programs. “We may have put people into that program who didn’t need to be there,” he said.

Several admissions officials and high school guidance counselors said that the SAT score situation has been the topic of considerable discussion in the last week or so. Chiara Coletti, a spokeswoman for the College Board, said via e-mail that “not a significant number of colleges” have contacted the board about the issue. She said that nine colleges had alerted the board to declines, and that the board then decided to send out the memo “proactively” to admissions officers, and that not many had since expressed concern.

Coletti said that the College Board did not plan to notify test takers at this point of the concerns about the scoring decline, in keeping with the board’s approach of not sharing incomplete cohort results. Some students might find the information on the College Board Web site, she said. It appears there under the headline “206 SAT Cohort Averages” on the portion of the site for educators. Admissions officials said that applicants appeared unaware of the issue. “Telling students specifically about average scores to date for a part of their cohort wouldn’t seem to serve them,” Coletti said.

— Scott Jaschik

Want it on paper? Print this page.
Know someone who’d be interested? Forward this story.
Want to stay informed? Sign up for free daily news e-mail.

Comments

The ACT advice worked well

I posted a comment in 2006. Since then I had my son take the ACT. He scored 25 which was used in all of his college admission applications. His ACT score was considered higher than his SAT which was 1050 (Reading and Math). He is now a freshman at Cal Poly. I would advise anyone to have their child take both tests. I have advised others to do this and all have experienced better results with the ACT score. You need every advantage when applying for college. Good luck to all!

Dan Wishard, Architect, at 5:45 am EST on November 27, 2007

Why is there a knee-jerk reaction that something is suddenly wrong with the test rather than something is suddenly right?

david stong, at 6:00 am EDT on May 3, 2006

Settle an old score

Why admissions employees rely on SAT – it does not level the playing field.‘coaching, cost, and test preparation give a distinct advantage to the affluent students’

Why admissions employees ignore prospective students’ high school records? It is a proven track record – it is enormous – that is accumulated in years – they choose to rely on a mere number that is issued in a three hours on Saturday – by a profit seeking company?

It is butchery – to ignore the high schools records – it is like ignoring Barry Bonds homerun record because in a game he did not hit a homerun that would be in morons’ world – that is what admissions people do – especially toward African Americans.

Imagine what would admission idiot tell Barry Bonds? No homerun – so you need to be enrolled in an enrichment program – this has been admitted by a La Salle University incompetent – how many students were forced to unnecessary and unwarranted enrichment programs? – I assume all of them are African Americans

I believe admissions employees are guilty as college board – thus they should be sued jointly.

David Robertson, Professor at SUNY, at 8:30 am EDT on May 3, 2006

New SAT

The problem we have, if the new SAT is scoring lower than was expected, is that colleges on rolling admission made admissions and merit scholarship decisions based on the belief that the scores were the same as last year. No one is commenting on the validity or lack of validity of the test. We just made decisions based on an assumption that doesn’t appear to be correct. Is that our fault? Maybe? If the overall drop in scores is 4-5 points, there are some who must have scored 20 -30 lower and some who scored higher. On balance, the scores are lower. Some students lost institutionally-awarded scholarships and and received lower need-based packages as well.

Bob Voss, at 8:40 am EDT on May 3, 2006

New SAT mystery and Barry Bonds????

I’ve read a lot of pretty wacky postings on this site. Comparing Barry Bonds home run record and baseball career to a college selection process is the stupidest analogy I’ve ever read...and this from a college professor.

Wonder if he has tenure?

feudi@earthlink.net, at 8:50 am EDT on May 3, 2006

The College Board Won’t Admit the Problem

There is something awry here and The College Board doesn’t want to admit it. They are experiencing the kickback of unraveling this new exam too quickly without adequate testing and analysis. While I hope they analyze the problems admissions offices and high schools are experiencing in declining score results and publish findings in line with this, it is likely that they produce a set of numbers that ‘dismisses’ the problem.

Admissions Officer, at 8:50 am EDT on May 3, 2006

SAT scores

If SAT scores drop while high school GPAs remain as high as ever, it could indicate that our nation’s grade inflation is catching up with us. Not every college applicant is a true A or B student in all subjects.

Ree, Teacher, at 9:20 am EDT on May 3, 2006

It seems a little strange to look at the scores institution by institution, rather than overall, in order to attempt statistical analysis. If students’ scores overall dropped discernibly, that means something about the knowledge of the students taking the test, the population of students taking it, the test questions themselves, general test design, test administration, test scoring, or test reporting (other variables may have slipped by my notice). Institution-by-institution score averaging introduces other possibilities, though.

I’ve read a fair amount in the last year about students applying to more and more colleges. If this practice has indeed become more widespread, and in particular if it has become more widespread among students below the top percentiles, it could explain a drop of a few points in average application scores at institutions nationwide pretty easily.

I’m not generally a big defender of standardized tests, but it seems like there’s not enough information here even to say there’s been any actual deviation from the past as far as the test itself is concerned.

Thane Doss, Tokyo

Thane Doss, at 9:50 am EDT on May 3, 2006

grade inflation

Ree’s comment seems very logical. For years faculty at both the high school and college levels have worried about grade inflation. The number of 4.0 averages have risen to the point that some colleges have turned to class rank. Though shocking to some commentators, the possiblity that the SAT has documented the flaw in the current education system is very real. What are the effects if the profession takes the score drop as real and the gradepoints and rank as faulty? Worth serious thought.

Kurt, at 10:15 am EDT on May 3, 2006

Validity: Predictive or fiduciary?

This blatent SAT scoring error and the consequences for all of the examinees serves as further evidence of the larger validity and reliability issues at stake for American students and their families. For many years we have known that the predictive validity of the SAT is the mathematical equivalent of a coin toss, yet this economic monster continues to breathe fear into the hearts of examinees as it greases the wallet of ETS. This is a meager complaint in comparison to the damage that is inflicted upon the millions of Hispanic and African American youth who continue to score one standard deviation below the mean compared to their white brothers and sisters. The long-term implications for these individuals is immeasurable. Where is ethic’s role placed in this decision-making when all that seems to be considered is the fiduciary validity of this cash cow. Fair test? This test reflects capitalism at its best and democracy at its worst. Can we for once apply best practices in the business of education and stop misapplications of science in assessment?

Greg S. Goodman, Dr. at Clarion University, at 10:40 am EDT on May 3, 2006

Why not do away with scores and use something more objective: take in students based on tutition their parents willing to pay; their atheletic abilities; and sworn contact with the school not question authority?

nate zuckerman, Adjunct retiree, at 12:05 pm EDT on May 3, 2006

Grade inflation is not the problem—- —-If grade inflation is a reality, it has been taking place over a period of time. There is no reason to think that grades would take a huge jump in one year at most of the schools in the country. —-Notice that the article does not mention GPA. It only mentions class rank and courses taken. You have to assume that if there is grade inflation that it is affecting all students at a school. Therefore, grade inflation does not affect rank.—-This score drop took place in the same year that a much modified, hastily constructed, and inadequately pre-tested new SAT was released. This is more than a coincidence.

One possibility is that the test prep companies have not figured out and taught their customers all of the tricks yet, as they did on the old test. As soon as they do, scores on this highly coachable test might start climbing.

Most likely though, the cause of the problem is another set of errors done by companies that can get away with anything they want without oversight and outside scrutiny.

Brad MacGowan, high school counselor, at 1:05 pm EDT on May 3, 2006

Fair Test

Why not give a test that has strong reliability and validity measurements such as the Stanford-Binet, the WAIS-R or any other measurement with heavy g loadings?

thomassowellfan, at 1:25 pm EDT on May 3, 2006

Test Fatigue?

Perhaps the fact that the SAT is almost an hour longer because of the addition of the writing section play a part in the decline of scores. Test takers most likely experience a decline in concentration, reasoning and analytical ability the longer they have to sit for a three plus hour test with one or two five minute bathroom breaks.

Dee, at 2:05 pm EDT on May 3, 2006

Big Problems?

I fail to see what the immediate damage to college applicants might be, other than those schools who do make a quick, shoot-from-the-hip decision to offer early admissions.

It might be possible for those schools on a rolling admission schedule to fill up all their spots, but that would seem illogical.

If a student’s class rank, high school GPA, and SAT are all three brought down by the lower SAT scores (HS class rank and GPA aren’t subject to much fluctuation), it would seem that all college applicants would be on the same admissions footing.

Schools with a rolling admissions program, should only fill up all their freshman seats more quickly if all applicants’ rank/GPA/SAT ratio were suddenly HIGHER, not lower.

Critiques of the rank/GPA/SAT ratio consideration for admissions don’t seem to mind if colleges waste new students’ time and money, only that the cause of supposed egalitarianism is served.

I’ve yet to read or hear about any institution of higher education that uses only the applicant’s SAT or ACT score. In the present world, pulse/98.6 body temp. and ability to pay are the most important criteria; unless that student can catch a wobbly, wide pass in a mass of defensive backs and run well.

Dr. F. Gump, at 2:20 pm EDT on May 3, 2006

“They are experiencing the kickback of unraveling this new exam too quickly without adequate testing and analysis. “

Yep. And the colleges themselves are responsible for the results. They could have refused to accept the new SAT, which was never fieldtested in its final form. Instead, they shrugged and accepted it.

My guess: the decline, assuming it exists, is caused in part by the fact that the test is shorter and easier, particularly the math section.

I wrote it up in more detail here.

Cal Lanier, at 4:25 pm EDT on May 3, 2006

Professor Robertson asks why not rely on high school records — class rank and grades in essence rather than standardized tests. The answer is obvious: comparison of high school records between schools in different states, within states, and even sometimes within school districts is not reliable.

The fundamental appeal of standardized testing (other than the ease of scoring) is that by testing applicants in Big Stone Gap, Virginai, and those in San Francisco, California, on the same material provides an opportunity for those whose school’s quality is unknown to the admissions office with the opportunity to demonstrate they are good candidates for admission.

High school quality varies enormously across the country on many bases: funding, quality of faculty, student body quality (affluence?), courses offered, etc. To rely on high school records without both extensive and intensive research into each school is to discriminate unfairly against those who come from particularly good and competitive schools in favor of those who come from poor schools. A valedictorian from a small high school in the middle of Nevada, or from an urban ghetto high school, may have a 4.0 average, but not be remotely as well-prepared or as well-qualified as a candidate from the middle of his or her class at an elite public high school (or private preparatory school), where there might be as many as 50 national merit semifinalists and commended students. In theory, a standardized test will provide a way of comparing the two that their records will not.

CatoRenasci, Apples and Oranges, at 11:40 am EDT on May 5, 2006

Cal Lanier’s comment that the new SAT was never field tested is stunning. It is hard to accept that a testing program of this cost, magnitude, and importance would not have been extensively tested in every creditable way.

But this is not the first SAT controversy. A much earlier renorming also raised questions concerning accuracy and need. As a result, the nation lacks the ability to do stable, long term comparisons of student performance. It is unfortunate.

Marvin McConoughey, at 5:35 am EDT on May 7, 2006

Don’t be so quick to discount Grade Inflation

I am well aware that the subject of “Grade inflation” has been cycling hot and cold for many years but I think it unwise to discount it simply as a matter of course. The strange thing to me is that the notion of grade inflation does not cycle “cold” because someone does high-quality empirical research that shows it doesn’t exist but rather because anyone who thinks it (grade inflation) exists is branded a heretic. Most individuals in a position to fund research in this area choose to (apparently) simply maintain the status quo rather than support efforts to learn the truth. In this case, and as Brad MacGowan pointed out, perhaps the students simply have not learned the “tricks” for the new assessment. As such, provided that the two forms are statistically equivalent, this disparity may actually be a good indicator of just how rampant/widespread grade inflation really is. Once students learn the new “tricks", their SAT scores will again be artificially exaggerated to match their grade-inflated rankings/performance in grade school. If you, as a teacher/professor, give the same exam semester after semester then during one semester you create a new exam, do you immediately decide the new exam is “bad” when the average grade in the class drops a letter grade or more? Do you even bother to look for other explanations? After the SAT “coaches” have picked the new assessment apart and taught the “tricks” and at the point of SAT and High School Performance re-convergence, everyone can once again wipe their brows, pat each other on the back, hand out some awards during celebratory luncheons, and go back to laughing about all of the people who believe in the theory of Atlantis, UFOs, the Loch Ness Monster, and Grade Inflation. Absolute insanity ... wake up, people; grade inflation at all levels (K-20) is slowly killing us and will continue to do so even if we ignore it. Do you think they have grade inflation in Singapore, Korea, or Japan? How about in Belgium, The Netherlands, or Switzerland? I would seriously doubt it and their world rankings in education – where they are kicking our backsides — reflect that fact. Grade inflation is not the only problem but it is certainly a serious symptom of something much more systemic. We (the United States) are the richest, most powerful, dynamic nation in the World but we can’t break into the top 20 internationally in science and math? What’s wrong with this picture? Wait! I think I just saw Nessie!

Alan Lott, Professor, at 10:25 am EDT on May 8, 2006

SAT scores down — maybe they’re accurate!

As a parent of a Senior in high school and an admission’s professional, I say these students who were the ‘whole reading’ method of reading kids, cannot read as well as they should and it’s showing up in the test scores. In addition, a lot of kids had not taken precalculus before taking the test or we a month or two into it if they did take it senior year, another valid reason for a decline. And, we all know that the majority of students have a difficult time just sitting down and writing an essay, so I say, of course the scores are likely to be lower! We need to have some data as feedback to high schools so they know how to adjust their curriculum. This shouldn’t be about what we report to US News & World Report.Rosemary Kelly

Rosemary Kelly, at 3:15 pm EDT on May 10, 2006

Maybe the ACT is an option

To all those in the admissions community: Perhaps with all the negative discussion herein we should begin to think about moving to the only curriculum based, college readiness, and college admissions test that is linked to the academic learning standards in 40 of the 50 states — The ACT.

Jennifer Sandata, Private Counselor, at 11:50 am EDT on May 11, 2006

Something Must Be Wrong

My son is in 11th grade and has taken the SAT for his second time. Each time he thinks he has done well, until we get the score reports. We have been told by admission counselors to look at only the reading and math portions. He is scoring in the 1500’s with reading and math in the 1020&1050 total. This is short of the 1200-1400 he will need. He has friends that are seniors that have taken both the old and new SAT. They are reporting similar scores in this years test when last year they had scores of 1200-1400. How could they drop that much? These are kids that are enrolled in upper level AP and Honors classes? Something must be wrong!

Dan Wishard, Architect, at 10:25 am EDT on May 12, 2006

SAT Validity

A couple of previous posters have noted the ’shorter’ math section and yet the overall longer test with inclusion of writing. Two things here: Even though the Math is shorter in terms of questions, that means that fewer number of errors are needed to drop a score significantly. Additionally, it’s not a 3+ hour test. Try 4.5+. The test centers call for your kid to be there an hour early (745am), and then don’t actually start at 845, with MINIMAL if any breaks for food or water until dismissed usually around 130-2 pm. This is our experience here and believe that the total experience is obviously having a negative impact on our students’ performance. I think the college board has literally killed the golden goose with this new version and this current 3 part monstrosity of test is not long for the world.

Jon, at 1:30 pm EDT on May 16, 2006

Race and SAT Scores

This is in response to the poster who on May 3 commented on the perceived racial issues regarding the SAT. I’ve heard these arguments before centering on cultural and economic disparities and I must respectfully disagree.

When I took the test, 18 years ago, I scored quite high (1200) without any special tutoring or expensive program. Since I came from a lower middle class socio-economic group within the City of Philadelphia I fail to see the advantage I had due to the fact that I am Caucasian. My parents did not progress beyond high school, my friends parents, neither.

What my parents did, however, was insist upon dedication to academic achievement, and I do not believe that to be a racial or economic issue. Rather, it is an issue of personal responsibility. My child now goes to the same public school that other ethnic groups have access to and goes to the same library that they have access to.

I truly do not understand the perceived racial disadvantage. Additionally, I think that people do a disservice when they espouse the idea of racial disadvantage. To put it another way, it would be the ultimate in arrogance to suggest that I performed better due to my race. I believe that the idea of racial advantage/disadvantage only serves to alienate the races. The idea that outcomes are dependent upon race is culturally and racially arrogant. It teaches those in the minority a “Why bother?” attitude and smacks of snobbery and elitism.

Brenda Mayer, at 2:05 pm EDT on May 28, 2006

could it be that we are beginning to see the effects of an over pressured, striving too hard, doing too much stupid community work, bunch of children exploding slowly.

The SAT magicians who we never see or hear who live in their own protected franchise should be given detention for adding to our childrens angst.

zalman shenk, professor, at 6:50 pm EDT on August 14, 2006

No SAT declines? Now what?

Interesting how many people had an explanation for the apparent SAT score declines this past year. It now looks as if (we’ll find out in a couple of days when the College Board releases scores) there was no decline. Admissions’ officers who “noticed” downward trends scrambled to find other colleges that fit that pattern. Critics of standardized tests fired salvos. Parents whose children did poorlysupplied anecdotes about other kids whose scores had dropped thereby “establishing” some supposedly hard and fast evidence. What this false alarm may really point to (if it does turn out to be a false alarm) is the degree to which we are all capable of misreading situations by seeking out evidence which merely confirms our preexisting beliefs, a phenomenon known as confirmation bias.

Patrick Mattimore, Teacher, at 11:55 am EDT on August 23, 2006

Jobs Related to New SAT Mystery

or search for jobs directly.

Assistant Professor, Sociology
Appalachian State University

The Department of Sociology at Appalachian State University invites applications for a tenure-track nine-month faculty ... see job

Executive Assistant REPOSTED
Princeton University

Position Summary: University Health Services (UHS), located in the McCosh Health Center, is a fully ... see job

Physician Faculty — Ftmo, Clinical Asst or Assoc Professor
East Carolina University

East Carolina University, a constituent institution of the University of North Carolina, is a doctoral institution with an ... see job

Tenure Track Assistant Professor Early Childhood and Special Education
West Chester University of Pennsylvania

Join a vibrant campus community whose excellence is reflected in its diversity and student success. West Chester ... see job

Assistant Professor, Education Research
College of the Bahamas

DUTIES AND RESPONSIBILITIES Candidate should have a Ph.D. or Ed.D. in Education Research with a minimum of 3 years of ... see job

Assistant Professor of Economics, Tenure-Line (Economics & Finance)
Bard College

The Economics Program of Bard College invites applications for a full-time tenure-line position specializing in financial ... see job

Adjunct Faculty Credit — Speech
Harper College

Job Description: Teach Introductory Speech courses (SPE 101) to our undergraduate population.

see job

Research Nurse
University of Minnesota, Twin Cities

The University of Minnesota is a premier employer and a talent magnet attracting leading faculty and staff from around the ... see job

Assistant Professor in Industrial and Organizational Psychology
University of Minnesota, Twin Cities