Passing schools, struggling students: Colorado reconsiders its formula for rating schools

The vast majority of Colorado schools and districts get a passing score from state regulators who track their performance. Yet fewer than half of Colorado third-graders meet state expectations in literacy and just 34 percent meet state expectations in math.

This disconnect has members of the Colorado State Board of Education calling for a change in how much weight the state gives to certain factors in determining whether a school or district is doing its job or needs more oversight.

“Both of those things cannot be true,” board member Steve Durham, a Colorado Springs Republican, said Wednesday. “You cannot characterize the student as not performing and the school as performing.”

Colorado’s school accountability system rates districts based on achievement on state literacy, math, and science tests, on annual academic growth, and, for high schools, on postsecondary readiness as measured by graduation rates, dropout rates, scores on college entrance exams, and enrollment in college.

Schools rated in the lowest two tiers – turnaround or priority improvement — go on the state’s performance watch or “on the clock.”  Such schools face state intervention, which can include closing or turning over management to a charter organization, if they don’t move into a higher tier after five years. So far, the state has shied away from drastic action and approved improvement plans brought forward by districts themselves, but that could change as some have not shown enough progress.

At the elementary and middle school level, 60 percent of a school’s rating is based on growth, a measure of how much progress students make compared to other, similarly situated students, while 40 percent is based on achievement, a measure of what students know. At the high school level, 40 percent of a school’s rating is based on growth, 30 percent on achievement and 30 percent on measures of postsecondary readiness.

Achievement on standardized tests is strongly correlated with students’ socioeconomic background, and many experts believe growth scores are a better reflection of whether schools are helping students learn. 

But there remains the troubling question of whether students are learning what they need to know in school, whether that’s third-graders having the literacy skills to carry them through the rest of their education or high school graduates being able to attend college without taking remedial courses.

“Is it time for us to put greater weight on achievement, since that is where we want to go?” asked Angelika Schroeder, the Boulder Democrat who chairs the state board. “We want to see growth, but achievement is what matters.”

The ratings formula ties into a long-running debate among testing experts. Groups like The Education Trust, which supports test-based accountability, argue that growth models water down expectations for disadvantaged students and don’t measure whether students will eventually reach proficiency. Others argue that achievement data is too closely tied to poverty to be a meaningful measure of school performance.  

In Denver Public Schools, parents and civil rights groups have questioned how schools could be rated green based on growth rates when most students in those schools couldn’t read on grade level. The district continues to tweak its own school performance framework in response to criticism.

Some states also use a hybrid measure known as growth-to-standard that looks at how long it would take students to reach grade level if they continued to make progress at the same rate.

This measure comes in for some of the same criticism as achievement data.

“As an accountability metric, growth-to-proficiency is a terrible idea for the same reason that achievement-level metrics are a bad idea — it is just about poverty,” Cory Koedel, an economist at the University of Missouri who has studied school accountability, told Chalkbeat last year.

Colorado’s school ratings used to include a growth-to-standard measure as a major component, but they haven’t taken it into account since 2015, when changes in assessments made year-over-year comparisons difficult.

Now that the assessments have stabilized and comparisons are more appropriate, the state will be adding growth-to-standard back in, as required by law. That provides an opportunity to revisit how much weight each factor in the school performance framework gets. A technical advisory panel will be studying the issue this fall and make recommendations to the state board.

One of the questions state board members want answered before they render a decision in early 2019 is how applying a standard that more heavily weights achievement would have affected school ratings and the possibility of state intervention.

Preliminary ratings based on 2018 test data placed 90 percent of Colorado school districts and 83 percent of schools in the higher tiers that essentially leave the schools free to do their work as they see fit.