[E]very spring, thousands of Vermont students sit down to take computerized tests in math and English. The tests, created by the Smarter Balanced Assessment Consortium, are administered to children in grades 3, 4, 5, 6, 7, 8 and 9.
Statewide results for 2018 were released in the fall, nearly a year ago, and scores on most tests ticked upward. More than 50% of students scored proficient or higher on English tests. Students scored between 35% and 52% proficient in math.
Data collection problems at the Vermont Agency of Education delayed the release of school-by-school results by more than nine months.
State officials finally published local data in August. The raw data, which we used for these interactive graphics, is available online.
Marginalized students – low-income students, students with special education needs, and racial and ethnic minorities – tend to score much lower on standardized tests. This phenomenon is so persistent — nationally as in Vermont — that it has a name: the achievement gap.
That’s why we built our interactive graphs to allow people to see how similarly situated students fared across different schools. Using the toggles, you can see the average score on any particular test at each school, but you can also compare how different types of students (say, low-income, or students on a special education plan) performed across schools.
Here’s a graphic that allows you to take a look at how low-income kids performed as a cohort in each school, as well as their more affluent peers. (It also allows you to look at top-line performance.)
And here is an interactive that allows you to do the same, but for students with or without special education plans:
A few notes about how to read the graphs:
The test’s algorithm uses the information to assign the student’s performance a number, called a scale score or scaled score, on a range from 2000 to 3000.
Students’ SBAC scores are graded on a scale of 1 to 4. Students scoring at achievement level 1 are not meeting expectations. Students who have mastered the material for their grade score at level 3, which means they are proficient, and if they score at level 4 they are advanced. That is why we report results as proficient and above. So-called “cut” scores delineate the scale score at which a student has demonstrated proficiency.
Sometimes, schools will disappear from our interactive graphs. That’s because if fewer than 11 students in a particular cohort took a test in any given grade, the results aren’t reported publicly by the Agency of Education in order to protect student privacy.
It’s worth noting that education officials often complain that standardized tests are too reductive a measure of school quality, and that results mostly reflect a school’s demographics.
“If we’re only looking at test scores, that’s too narrow a base by which to judge a school by,” said Bill Mathis, a member of Vermont’s State Board of Education, who is also the managing director at the National Education Policy Center at the University of Colorado Boulder.
The public will soon have a wider range of indicators with which to gauge school quality. The state has developed the Annual Snapshot, an interactive dashboard, to provide school-by-school information about personalization, cost-effectiveness, climate, and staffing.
But data problems have slowed that process as well, and only data on academic proficiency – i.e., test scores – are presently available on the dashboard. State board members, including Mathis, have complained about this, and agency officials say the Snapshot will be updated with additional data later in September.
Another thing to keep in mind: The intensity of student needs within a district often drags down performance. That’s why state and federal education funding formulas often apportion extra money to schools when a certain proportion of their children are low-income.
“Concentrations matter a great deal,” Mathis said.
And indeed, one of the consistently lowest-performing school districts in the state on SBACs is Winooski, a district with significantly higher rates of poverty, special education needs, and English-language learners than the rest of the state.
Winooski Superintendent Sean McMannon, in an email, said low test scores in the district can be attributed “primarily to the complex needs of our students.” About 70% of the district’s students qualify for free and reduced lunch, a key indicator of poverty, 41% qualify for English language learning supports, and 25% are on a special education plan.
But McMannon also added that that test scores show that Winooski students make steady progress over time. Internal district data shows that about 70% of students make at least one year’s worth of growth in math and reading each year, he said. State data reflects this as well: Vermont’s Snapshot shows that the district is “excelling” according to the dashboard’s performance change indicator – a benchmark for how well students improve over time.
Many of the state’s highest-performing districts, meanwhile, generally include Vermont’s more affluent schools. South Burlington High’s students, for example, scored in the top five in both math and English. And 62 percent of the school’s low-income students in grade 9 were considered proficient or above in language arts – better than their peers in any other school. But only 18% of the school’s students are on free and reduced lunch, well below the statewide average of 38%.
South Burlington High School Principal Patrick Burke said a few things contribute to the school’s success: a strong middle school program (the district’s middle school, Frederick H. Tuttle, is a high-performer on SBACs), no tracking in English in the first year of high school (which means certain kids aren’t put in lower-level classes), and a well-articulated curriculum.
The district also pays its teachers handsomely – faculty members made, on average, $80,600 that year – over $20,000 more than the average Vermont teacher’s salary. That’s in line with an analysis performed last year by VTDigger, which found that higher-performing districts tended to pay their teachers better.
Burke said the higher pay scale certainly helps keep high-quality teachers on staff, and turnover low, which means better follow-through on initiatives.
“Having stability within the faculty is definitely helpful,” he said.
Burke also suspects the school’s access to transportation helps keep kids engaged. Because the school is by a bus line, that gives more students opportunities to pursue opportunities in the community alongside their schoolwork.
“That’s just geographic. It’s not really something that we should break our arms patting ourselves on the back for,” he said.
We also crunched the numbers on how performance on tests changed over time. (Testing took place in grade 9 for the first time that year, so comparisons to previous years aren’t possible):
And finally, we looked at how performance was impacted by race and ethnicity. We didn’t include a school-by-school breakdown here, because too few schools have large enough cohorts to report out publicly:
