The number of Vermont students scoring proficient on the statewide test dipped this year, and a stubborn gap remains between students of means and those struggling with poverty.
State education officials said they do not know why scores went down.
“We were disappointed to see those score declines,” said Education Secretary Rebecca Holcombe. She attributed any narrowing of the achievement gap between low-income students and their wealthier peers to a drop in the latter group’s progress, not a stronger performance by the lower-income students.
“The achievement gaps between our vulnerable youth and students with greater privilege remain, and in some cases were narrowed, but this was largely a result of score declines for more privileged groups,” Holcombe said.
The Smarter Balanced Assessment Consortium tests students on Common Core state standards for English and mathematics in grades three through eight as well as grade 11. The threshold for proficiency is set high, according to the Agency for Education. The results help drive federal dollars to students who need the most support.
“These standards are deliberately ambitious, to ensure we have high expectations for our students,” the agency said in a news release. “Over time, the results will provide community members, teachers and parents with an increasingly reliable and accurate snapshot of children’s mastery of these standards as well as the progress of our schools.”
On Wednesday the agency released the statewide results for the spring of 2017. This is the third year the test has been administered, but 2016’s scores were used as the baseline, not those from 2015. Haley Jones, a spokesperson for the Education Agency, said last year that 2017 results will allow for comparisons.
“Next year there will be three years of results to look back on. Students will be used to the test, and we will get a better understanding of how they are doing,” she told VTDigger in 2016.
2015 was the first year for the test. That year, students took the exam entirely online. The SBAC is a computer-adapted test that changes the questions based on the student’s answers to get a more accurate account of the student’s mastery of the subject.
The 2016 test results are the baseline and this is the first year you can compare data, according to Jones. “Students have taken the test for three years with two years of back-to-back data so you could compare last year to this year,” she said.
But, she added, “it is up to you to compare.” The agency also does not rank schools. “You can do your own analysis,” she said.
What the agency did report Wednesday was not just the percentage of proficient students for each grade but the average scale scores — adjusted raw scores. This is important because the percentages can be misleading; a school with a higher percentage of kids performing at or above proficiency than another school could actually have average scale scores that are almost identical, according to Testing Director Michael Hock.
This year, 52 percent of third-graders scored proficient in mathematics. The scale score for proficiency was 2436, and on average Vermont students’ scale score was 2438.
Last year, 56 percent of students scored proficient in third-grade math. The statewide average scale score was 2437, one point less than what this year’s students achieved on average.
“It might look like a big change in terms of the percentage, but if you look at the scale score it might be they aren’t that big at all. Scale scores give a more accurate account of how those students and those schools did,” Hock said.
Last year, more than half of students scored proficient or above in English, meaning they met or exceeded the standards tested. This year, all but third grade saw more than half the students scoring proficient or above:
• 2016: 54 percent of Vermont students in grade three.
• 2017: 49 percent.
• 2016: 54 percent in fourth grade.
• 2017: 59 percent.
• 2016: 58 percent in fifth grade.
• 2017: 55 percent.
• 2016: 56 percent in sixth grade.
• 2017: 52 percent.
• 2016: 58 percent in seventh grade.
• 2017: 55 percent.
• 2016: 59 percent in eighth grade.
• 2017: 55 percent.
• 2016: 57 percent in 11th grade (the only year that the percentage dropped, from last year’s average of 58 percent).
• 2017 59 percent.
In mathematics, the percentages meeting or exceeding the standard were somewhat lower than last year:
• 2016: 56 percent of Vermont students in third grade.
• 2017: 52 percent.
• 2016: 50 percent in fourth grade.
• 2017: 47 percent.
• 2016: 43 percent in fifth grade.
• 2017: 42 percent.
• 2016: 41 percent in sixth grade.
• 2017: 39 percent.
• 2016: 46 percent in seventh grade.
• 2017: 44 percent.
• 2016: 44 percent in eighth grade.
• 2017: 41 percent.
• 2016: 38 percent in 11th grade (a 1 percentage point increase from 2015.)
• 2017: 37 percent.
Among the subgroups, more students of Asian descent scored at proficient and above than the state average across subjects and grades as they did last year, except in English in high school, where Asian students scored 3 percentage points below the statewide average. Blacks and American Indians, including Alaskan Natives, scored below the state average for proficiency across subjects and grades, as they had in 2016.
Vermont’s Hispanic students scored slightly below the state average for proficiency and above in both subjects in all grades, except high school math where they beat the state’s average by 3 percentage points. Gains made by this subgroup in English in grades six and eight were gone.
Last year, Hispanic students outscored the state average (56 percent) on English in grade six with 60 percent scoring proficient or better, but in 2017 only 51 percent hit the target. In grade eight in 2016, 68 percent of these students scored proficient on English (as opposed to 59 percent state average) and this year, 45 percent made it while 55 percent of all students scored proficient.
Students for whom English is not their native language performed worse than the state average in both subjects and increasingly worse in English as they advanced through school. By high school, only 1 percent of this group scored proficient or above in either subject.
White students tracked very closely with the state average. Girls did better than boys in English across all grades. They also scored close to boys, but did not do as well in mathematics, except in grades six, eight and and 11, where girls scored slightly higher in math.
Low-income students performed below the state average for proficiency in both subjects in all grades in 2017. In 2016, low income students performed just as poorly and did the worst in 11th-grade math, where only 19 percent reached proficiency. This year, that number dropped and only 17 percent reached it.
Holcombe said schools need to continue to engage all students as they put in place state initiatives such as personalized learning plans, flexible pathways, and early college. “We need to support our schools and teachers as they figure out how to support better learning outcomes,” she added.
State and school level data are available here.