Thank you for your interest in and support of the Valley News. So far, we have raised 80% of the funds required to host journalists Claire Potter and Alex Driehaus for their one-year placements in the Upper Valley through Report for America, a national service program that boosts local news by harnessing community support.

Please consider donating to this effort.

Vermont Test Scores Slide

  • Randolph Union High School ninth-graders gather for their daily advisory group meeting led by media center librarian Michelle Holder in Randolph, Vt., on Sept. 16, 2017. The 35 advisory groups of about a dozen students each span the school's 7-12 grades. Each group is together through their school career, which officials believe has helped to not only raise test scores but has also increased the graduation rate since advisories began with the class of 2015. (Valley News - Geoff Hansen) Copyright Valley News. May not be reprinted or used online without permission. Send requests to Valley News photographs — Geoff Hansen

  • Randolph Union High School teacher Josh Hester-Reyes reads a homework problem before going over the answers with juniors and seniors in the Advanced Placement Calculus class at Randolph Union High School in Randolph, Vt., on Sept. 15, 2017. Vermont's Smarter Balanced Assessment Consortium (SBAC) tests for 11th graders are designed to determine their mastery of academic skills needed for entry-level college courses and the workforce. (Valley News - Geoff Hansen) Copyright Valley News. May not be reprinted or used online without permission. Send requests to Valley News photographs — Geoff Hansen

  • Randolph Union High School junior Shea Fordham works to solve a problem in her Advanced Placement Calculus class at Randolph Union High School in Randolph, Vt., on Sept. 15, 2017. (Valley News - Geoff Hansen) Copyright Valley News. May not be reprinted or used online without permission. Send requests to

  • ">

    Rex, a character in the movie "Toy Story," is set on top of Randolph Union High School teacher Josh Hester-Reyes' digital projector in his classroom in Randolph, Vt., on Sept. 15, 2017. (Valley News - Geoff Hansen) Copyright Valley News. May not be reprinted or used online without permission. Send requests to Geoff Hansen

Valley News Staff Writer
Published: 9/18/2017 10:00:23 PM
Modified: 9/18/2017 10:00:28 PM

When Michael Hock learned how Vermont students scored on the statewide standardized test this year, even he was surprised — and he’s the director of assessment at the Vermont Agency of Education.

The results from the Smarter Balanced Assessment Consortium tests, taken by all Vermont public school students in grades three through eight and grade 11 in English language arts and math, show a decline in student performance from the 2015-2016 school year. Hock said this is an exception to the general rule that, as students and educators adjust to a new testing system, performance steadily improves over the next few years. This is the third year that the state has administered the Smarter Balanced test, and the second year it’s been considered an accurate measure of student abilities, said Vermont Secretary of Education Rebecca Holcombe.

“Frankly, we were as surprised as anyone,” Hock said during a phone interview last week, shortly after Holcombe released the performance data that reported a dip in the percentage of students who received a “passing” test score of proficient or above. He said one contributing factor to the overall decline in scores might be that schools are focusing on other issues, such as consolidating under Act 46 or creating more flexible learning pathways under Act 77 — issues that are not necessarily tied to test preparation, but which help shape the quality of a school’s learning environment nonetheless.

This is also the first year that scores will be used to hold schools accountable through the Every Student Succeeds Act, the federal public education law that replaced the landmark No Child Left Behind Act, by helping to determine the roughly 5 percent of schools that need support most urgently, Holcombe said in a phone interview last week.

No Child Left Behind, signed into law in 2002, required the annual standardized testing of public school students. This was not only to measure students’ progress in core academic concepts and to “grade” schools on how well they teach these concepts, but also to compare scores along such demographic lines as race, gender, family income and disability status. Closing this divide, which educators and policy-makers call the achievement gap, has proven a stubborn task for public schools.

The computerized Smarter Balanced test aims to evaluate students’ aptitude by finding the level at which the student is challenged, but capable: If they answer a question correctly, the next one is harder; if their answer is wrong, the next question is easier.

This year’s results did not completely defy expectations, though. In some ways, they confirm what is often the case: that more privileged students tend to score higher on standardized exams than low-income students, who are identified by their eligibility for free or reduced school lunch, or FRL.

This statewide performance gap did narrow somewhat since last year — a trend that Holcombe attributed more to lower scores among privileged groups than to higher scores among low-income students. Holcombe cautioned against giving too much weight to these results, comparing score declines to a “light on a dashboard” that doesn’t necessarily signal an imminent breakdown, but might warrant a look under the hood.

This light went off at a number of schools on the Vermont side of the Upper Valley, although, as Holcombe said, the signal is not without its blind spots.

For example, last year at South Royalton School, many low-income students did at least as well as their higher-income classmates, particularly in English and language arts, an outcome that Principal Dean Stearns guessed might be attributable to the school’s emphasis on writing, he told the Valley News at the time.

This year, however, was a different case: Some of the income-based proficiency gaps at South Royalton were among the highest on the Vermont side of the Upper Valley.

Two-thirds of the higher-income third graders, and less than one-third of their low-income peers, were proficient in English language arts. This disparity was even greater in fourth grade — 55 percent of wealthier students passed in English, compared to only 9 percent of the poorer students in the class.

In the 11th grade, 25 percent of low-income students were proficient in English language arts compared to around 62 percent of the better-off students, and only 8 percent were proficient in math compared to 46 percent of better-off students.

Reflecting on these results in an email exchange on Monday, Stearns said that given the lack of long-term data, “I struggle to come up with a reason for the drop in the FRL proficient percentages. All of our curriculum work and best instructional practices are still in place, and has been reinforced through professional development of the staff at all levels.

“All of these factors are not taken into account when the numbers are published by the AOE.”

He believes the most significant factors in tests like the Smarter Balanced assessment are also the factors that can be most difficult to measure: “the variability of the students, the motivation of the students, the frame of mind of the students on test day, how seriously the students take a ‘high stakes test’ that has no direct effect on their grades, etc.”

In this vein, Hock and Holcombe both cautioned against making snap assumptions about a school and its students based solely on the percentage of students who scored proficient or above, especially in small schools where outliers and invisible factors like the ones Stearns described might throw off the results.

Hock pointed out that proficiency-based statistics also do not take into account how close to the proficiency cutoff students were: Students may just scrape by one year, or fall just short another, so the change can seem more drastic than it really is.

“There could be a dog barking outside the window and so the class performs badly that year because the dog wouldn’t shut up,” Holcombe said. “What’s most powerful to me is that there is tremendous variability (of scores) among schools, even with the same composition of low-income students.”

Case in point: Whereas the performance gap at South Royalton widened in the 2016-2017 school year, Randolph schools saw some of the Upper Valley’s most comparable scores between FRL and non-FRL students.

At Randolph Elementary School, 66 percent of higher-income third-graders, and 63 percent of low-income third-graders, were proficient in math. And although seventh-grade math scores at Randolph Union High School were below the state average of 44 percent proficiency, the income-based performance gap was comparable, with 32 percent of low-income and 34 percent of higher-income students showing proficiency.

Low-income 11th-graders also outperformed wealthier students by 16 points in English and 7 points in math, though both income groups scored relatively poorly in the latter subject.

One exception to Randolph’s narrow performance gap was the 32-point gap in 8th grade math, which RUHS Co-Principal Dave Barnett said might have to do with staff turnover: He estimated that Randolph has seen five 8th-grade math teachers come and go over the past seven years.

He and RUHS Co-Principal Elijah Hawkes speculated that the largely comparable scores among income groups could be a byproduct of the school’s newer strategies for supporting more vulnerable students. Barnett said the percentage of low-income students at RUHS has doubled over the past decade, and that he and Hawkes have been trying to adapt to the changing needs of their students. Advisory groups reinforce close relationships with peers and educators, and project-based learning and workforce development at Randolph may be especially effective given the proximity of Vermont Technical College and the Randolph Technical Career Center, Barnett said.

Hawkes added that, in his opinion, providing students with opportunities to thrive at tasks “that have relevance to them and that they feel invested in personally” can translate to stronger performances on standardized tests, not just the Smarter Balanced assessment. Test scores give an incomplete picture of this student success, he said.

Hock, the state’s assessment director, echoed this sentiment, and questioned the role of test scores in encouraging schools to hold themselves accountable for meeting the needs of their students, particularly the ones that are most vulnerable.

“Frankly, I don’t think very many schools need the state to hold a stick over their heads to try to improve,” he said, adding that evaluating schools requires a more holistic perspective. To this end, the Agency of Education has started sending out teams to conduct “field visits” at Vermont schools, to get a more complete understanding of how schools are addressing the needs of more vulnerable students.

While test scores are one tool for measuring these improvements, Hock said, “you can’t built a house with just a saw.”

To view the results of Vermont’s Smarter Balanced Assessment Consortium for the 2016-2017 school year, visit and download the spreadsheet that’s linked on the page.

EmmaJean Holley can be reached at or 603-727-3216.

Valley News

24 Interchange Drive
West Lebanon, NH 03784


© 2021 Valley News
Terms & Conditions - Privacy Policy