This article by Erica Meltzer first appeared on Chalkbeat.org on December 4, 2020
Colorado shouldn’t use standardized tests to judge the performance of schools or teachers amid a pandemic, a group of education leaders and advocacy groups agreed.
But should the tests be given at all next spring? And if so, should the data be publicly released? The same working group couldn’t agree in a debate that speaks to long-running questions about what these tests are even for.
“Our assessments are designed for an education system that doesn’t exist” this year, Peter Hilts, superintendent of District 49 in Colorado Springs, said in a recent meeting. “We’re not measuring anything remotely like a consistent system.”
“The inconsistent conditions are the reason to administer the assessment if possible,” countered Luke Ragland, head of the conservative education group Ready Colorado, in the same meeting. “The lack of consistency in educational services is why we would want to measure” where students are.
Federal and state law require standardized tests — in Colorado, that’s CMAS for students in third through eighth grade and the PSAT and SAT for high school students — and describe how they should be used for school accountability and teacher evaluations. Any changes will ultimately require federal waivers as well as legislative action or an executive order from Governor Jared Polis.
Outgoing Education Secretary Betsy DeVos had indicated the US Department of Education was unlikely to grant testing waivers this school year. Whether to stick with that policy or give states more flexibility will be one of the first questions confronting incoming President Joe Biden and his education secretary. The decision not to administer the National Assessment of Educational Progress, often known as “the nation’s report card,” in pandemic conditions led some federal lawmakers to say state assessments are more important than ever this year.
The Colorado Department of Education convened a working group in September to make recommendations on many of the most pressing questions. The members, who included representatives of school districts, teachers unions, and advocacy organizations, reached agreement on several areas:
- High school students should still take the PSAT and SAT in the spring, assuming public health conditions allow it.
- The social studies test typically given to fourth and seventh graders and the science test typically given to fifth, eighth, and 11th graders should not be given this year.
- Any standardized tests shouldn’t be used for high-stakes accountability purposes, including teacher evaluations.
- Colorado should not issue school or district performance ratings next fall based on tests given in the spring.
The group could not agree, however, on whether the CMAS tests that measure reading, writing, and math skills should be given to third through eighth grade students or on whether the results of any testing should be made publicly available.
Jen Walmer of the Colorado chapter of Democrats for Education Reform said she talks to parents who are “in panic mode” about their children’s academic progress. The tests would provide data to take stock of learning loss. One administrator on the working group said she was struggling to reconcile her role as a district representative opposed to giving tests with her perspective as the parent of a student receiving “really uneven” online instruction.
Bret Miles, executive director of the Colorado Association of School Executives, a group that represents superintendents, said the disagreement boiled down to whether the assessments will yield accurate information.
“We’re at an impasse because of the different values we’ve placed on this assessment,” he said.
Colorado superintendents for months have called for the state to cancel the administration of CMAS, as well as other required assessments for young readers and English language learners.
Given the wide range of ways students are learning, the tests wouldn’t provide valid information, they argue, and administering them would be a logistical nightmare, whether students are at school or at home. Could remote students be compelled to come to school to take the tests? How many extra weeks would be spent in testing if districts adhere to social distancing? What if someone tests positive in the middle of testing, sending their classmates and teacher into quarantine?
Or if students are allowed to take the tests from home, how could you compare the results of a student who has their own room and laptop with those of a student working at the dining room table while she helps three younger siblings with their schoolwork or the student taking the test on a cell phone?
“Our overwhelming and energetic consensus is that state-administered and state-required assessments are counterproductive during the current health crisis and therefore we recommend that they be cancelled or made optional this year,” wrote 26 Pikes Peak region superintendents in a letter sent in mid October. They were later joined by Denver metro and rural superintendents.
Nicholas Martinez, of the advocacy group Transform Education Now, said in an interview that parents and the broader community need to confront what’s happening to students, particularly low-income students of color whose families have been more deeply affected by job losses, sickness, and death.
“Hiding that data, not presenting it publicly, is a disservice to those families,” he said.
Some parents may mistakenly believe their child is learning if they like their teacher or don’t balk at online learning, while others may be aware their child is struggling academically but mistakenly think it’s their own fault.
“When you see that it’s an entire system that failed, you can lean on your community and make changes,” he said.
Van Schoales, president of A Plus Colorado, an advocacy group that promotes the use of data to improve schools, said he strongly agrees with the recommendation not to use testing for high-stakes purposes, but he does want tests to be given.
“It’s really important that we do standardized testing so that we can better understand the impact of the pandemic and target resources, in terms of money and people and programs,” he said. “We’re completely blind in Colorado about this. It’s all anecdotal.”
In a recent survey, many superintendents reported that loss of reading skills among early elementary students was their top concern. Superintendents say they use internal assessments and diagnostic tools related to their curriculum to measure student learning, and that these tools, unlike CMAS, provide information they can use to help students within the current school year. But such data is not available to the public or easily comparable across schools or districts.
Derek Briggs, a professor at the University of Colorado Boulder and director of the Center for Assessment Design Research and Evaluation, describes himself as “certainly not anti-testing” — and as a researcher, he would love to have the data — but this year, in these circumstances, he sees more downsides than benefit to giving the CMAS tests.
Students typically take CMAS in March and April. With most Colorado students learning remotely and an uncertain timeline for returning to the classroom, many students will have just recently reconnected with their teachers when the testing window opens. That time would be better spent on actual teaching than test administration, he said.
“Is the ability to quantify the impact of COVID worth the cost?” he asked. The data itself would be difficult to interpret, he said, because there is so much variability, even within schools, in how much in-person learning students are getting. One third-grader might have months of uninterrupted schooling, while another had to quarantine three times. Even among students learning at home, some are in pods with private teachers while others are fending for themselves.
National test results released this week present many of the same challenges. Students did better than many experts had feared, losing some ground in math but largely holding steady on reading. But fewer students took the tests than usual, and some of the students who opted out might be those who struggled the most. It’s hard to say whether the results give an accurate picture.
Instead, Briggs suggests targeting resources like summer school and tutoring at the students we know struggled more, those with poor internet access and those whose schools already struggled to meet students’ needs.
“Most parents know they are in a suboptimal situation for learning, so why not operate under that assumption?” he said. “Do you need me to give you an assessment as a researcher to say that non-white students lost on average three months of learning and the white kids only lost 1½ months of learning? Do you need a test for that?”