Achievement gaps, shortcomings in math persist in latest SAT scores

The state Department of Education on Monday released the fourth year of results from its Connecticut SAT School Day, showing average scores for 2018-19 that were down just slightly from last year in English language arts and math.

"We would've hoped for it to be better than this," said Ajit Gopalakrishnan, the department's chief performance officer. "We always would've liked to see better performance, because as much as anybody else, we always want to see greater achievement" and for achievement gaps to close.

The Connecticut SAT School Day is the state's accountability assessment, in which the test is administered free to 11th grade students in public schools. The SAT was redesigned in 2016 to better align with Common Core standards, Gopalakrishnan said, and is "now more of a content-type test and not that arcane kind of vocabulary test."

The average ELA score this year — on the 200-800 scale — was 515, compared to the college and career readiness benchmark of 480. But the statewide math average was 501, lower than the benchmark score of 530. The respective ELA and math averages last year were 516 and 503.

Among local districts, only New London had an average score below the ELA benchmark. New London, Groton, Ledyard, Montville, Norwich Free Academy and Waterford all had average math scores below benchmark, while East Lyme, Lyme-Old Lyme, Marine Science Magnet High School, North Stonington and Stonington all scored above.

Marine Science Magnet students had the highest ELA scores in the region, with an average of 589, while Lyme-Old Lyme students had the highest math scores, at 580.

Gopalakrishnan said the College Board established the benchmarks from a study it conducted a few years ago, based on the 75 percent probability of students achieving a "C" or better their freshman year of college in the relevant subject.

For example, a student who scores at the 530 benchmark in math has a 75 percent chance of getting a "C" or better in a first-year college math course.

Gopalakrishnan said when analyzing results, it's important to consider that the share of students considered "high-needs" increased from 38 percent in 2015-16 to 43 percent this year. That category includes students who are English learners, in special education or from a low-income family.

"We still have a lot of work to do" on the gap between high-needs and non-high needs students, Gopalakrishnan said. The data also show persistent test performance gaps by race.

In each of the past four years, for both ELA and math, the lowest performing group was English learners, with special education, black and Hispanic students also performing below the statewide average. Asian students always had the highest average scores, followed by white students.

For example, the average math score for 2018-19 ranged from 375 for English learners to 426 for black students to 591 for Asian students.

Gopalakrishnan said that because of changing standards and changing assessments, he can't compare achievement gaps now to ones 10 years ago.

He maintains that the SAT is a good assessment for all students, and noted that test performance gaps are by no means unique to the SAT. Gopalakrishnan said there is no quick fix or shortcut to address achievement gaps, but that it's a combination of factors, such as good instruction, engaging students and class attendance.

Still, he stressed that the SAT is just one part of the overall picture, and pointed out that Connecticut is seeing higher graduation rates at struggling schools and more students who are college- and career-ready upon graduation.

e.moser@theday.com

Editor's Note: This version includes Marine Science Magnet High School specifically, instead of LEARN in general, and corrects which school had the top ELA scores in the region.

READER COMMENTS

Loading comments...
Hide Comments

TRENDING

PODCASTS