Photo

CreditJi Hyun Yu 

To the Editor:

Re “The Big Problem With the New SAT,” by Richard C. Atkinson and Saul Geiser (Op-Ed, May 5):

We applaud the writers for recognizing that the new SAT is a “more straightforward test of material that students encounter in the classroom.” However, their overall claim is outdated. It is no longer true that because the SAT provides normative score comparisons (to the performance of other students), it can’t measure student learning. The new SAT draws on advancements in assessment design such that classroom content now comes first, and students perform against that bar.

The new SAT unquestionably measures the knowledge and skills that research shows matter most for college readiness and success — precisely the work students are doing in classrooms every day. The assessment will provide students and teachers with specific feedback so students can improve their readiness for college. It will also remain a valued resource in college admissions.

To further advance equity, the College Board is working with the nonprofit Khan Academy, a leader in online education, to provide students everywhere with free, personalized practice tools. Through this unprecedented partnership, the College Board is leveling the practice field for all students.

CYNDIE SCHMEISER

Chief of Assessment

The College Board

New York

To the Editor:

Even with its coming revisions, the SAT remains “problematic” at best, as Richard C. Atkinson and Saul Geiser note.

A college admissions exam is supposed to predict academic performance accurately and fairly while resisting high-priced coaching. The SAT has long fallen short of these goals. The planned revisions do not address the test’s basic flaws.

The SAT will remain a weak predictor of undergraduate success. High school grades will continue to forecast college graduation chances more accurately than any test does. The exam will still under-predict the performance of young women, students whose home language is not English and older applicants. Well-to-do families will not stop buying their children test-prep “steroids.” SAT scores will remain a better measure of family income than college readiness.

The planned changes are largely cosmetic surgery, making the SAT look more like the rival ACT exam, which has become more popular.

Neither test, however, is necessary for evaluating applicants, as more than 850 test-optional colleges and universities across the United States regularly demonstrate.

ROBERT A. SCHAEFFER

Sanibel, Fla.

The writer is public education director at FairTest: The National Center for Fair and Open Testing.

To the Editor:

Perhaps the biggest problem with the new SAT is that it will be aligned with the Common Core standards, even though many students taking it will have had little to no exposure to the Common Core. States are all over the map in the timing of their implementation plans, with some beginning just this year and others pausing or even rejecting the standards.

The correct move would be for the College Board to delay this revision so that all students stand a fair chance.

LISA EGGERT LITVIN

Hastings-on-Hudson, N.Y.

To the Editor:

Richard C. Atkinson and Saul Geiser say the most significant problem with the new SAT is that it will “remain a ‘norm-referenced’ exam, designed primarily to rank students rather than measure what they actually know.”

Instead of this “bell curve” approach, the writers champion the K-12 model of “criterion-referenced” tests, which measure how much students know about a particular subject “in relation to fixed academic standards.”

After over a decade of standardized testing mandated by the federal No Child Left Behind Act, however, it is clear that “criterion-referenced” proficiency exams come with their own set of problems. Scores such as “needs improvement,” “proficient” and “advanced” are not terribly informative.

Moreover, where psychometricians or policy makers choose to draw the “proficiency” line is a subjective judgment call, which challenges the very notion of “fixed academic standards.” Indeed, nearly one-third of all states lowered the cutoff scores for their No Child Left Behind tests in order to raise the number of students deemed “proficient” in math or reading, a reprehensible kind of statistical chicanery that provided an illusion of academic progress.

If the new SAT follows the lead of K-12 testing, high school students would ultimately be sorted into two groups: those deemed college-ready and those deemed unprepared. The writers’ proposal would not make college admissions “fairer and more rational.” To the contrary, following whatever arbitrary definition of college-readiness the College Board settled on, it would result in telling thousands of students they simply are not “college material.”

JEFFREY AARON SNYDER

Northfield, Minn.

The writer is an assistant professor of educational studies at Carleton College.