While testing experts are optimistic amid high participation in some states, they urge caution in interpretation, particularly at the local level.
From K-12 Dive
By Naaz Modan
Oct. 29, 2021
Following a brief hiatus in statewide assessments, and shortly after Miguel Cardona took over as secretary of education, the U.S. Department of Education asked local and state education agencies to get back into the swing of federally mandated testing for the 2020-21 school year. As the results slowly trickle in, they show a patchwork of attendance and outcome levels nationwide.
“Variability is the watchword this year,” said Scott Marion, executive director at the Center for Assessment, a nonprofit focused on meaningful assessment and accountability practices.
While not all states have reported results, those that have show testing participation rates ranging from a low of 10% in New Mexico to a high of 97.5% — above the federal requirement of 95% — in Louisiana. In some cases, others have yet to report because they chose to test over the course of an extended fall timeline.
Overall, the participation rates are more promising than expected, according to Anne Hyslop, director of policy development for the nonprofit Alliance for Excellent Education.
“The data are really encouraging because I feel like there was a widespread assumption that it would be very difficult to assess students last spring given all the disruptions in classrooms due to COVID,” Hyslop said. “It turns out in many states that wasn’t the case.”
Data collected by the Center on Reinventing Public Education, which is tracking state assessment information and last updated its database at the beginning of October, shows at least 12 states had overall participation rates of more than 90%. Out of those, at least four met the 95% threshold traditionally required by federal law, but which was waived across the board by the Department of Education for the 2020-21 school year.
However, Dale Chu, assessment expert and independent education consultant, urges more caution even in states that are reporting high participation numbers.
“Even in states that had high participation rates — like Louisiana, Indiana, Tennessee — that’s great but it’s still very uneven,” Chu explained. Some of the “most neediest localities” in Tennessee, like Nashville and Memphis for example, had lower participation rates.
“Even within states that have relatively high — not like 95%, but 80-85% — you see some big districts in those states with 50% participation,” Marion added.
Why did some states fall short?
Other states reported much lower overall testing participation rates. Oregon, for example, reported 30%, while New Mexico hit a low of 10%.
This could partly reflect COVID-19 health and safety concerns, experts said, considering students in some states reporting lower numbers may have still been learning remotely. Lack of mitigation measures and lack of access — like limited testing sites, time frames, and transportation to and from schools — could have steered parents clear of assessments, Chu added.
“But I think in the extreme cases where you see very low participation rates, that reflects more a lack of political will to give the assessment than those logistical challenges,” Hyslop said.
States that reported low numbers are also more likely to have testing aversions to begin with, Chu agreed. “In states where there was strong leadership and commitment and thought that tests were one part of the overall recovery, you saw higher percentages there. And when state leadership didn’t believe that, you saw it on the other end.”
Though the Department of Education received a number of waiver requests, most were declined, according to a department spokesperson. Of the 50 states and the District of Columbia, the latter was the only to receive a blanket waiver for federally mandated testing due to a bulk of its student population still learning remotely. Colorado was also granted some flexibility.
How can the data be used?
Assessment scores still have something to offer despite variability in participation rates, testing experts agreed. In places where states had high participation, results can be used as they were before, said Christine Pitts, resident policy fellow at the Center on Reinventing Public Education.
“For states that are seeing lower participation rates, especially at the local level, I think it’s important that school district leaders still use the information,” Pitts said, adding they should be only one piece of the larger puzzle.
Districts can also use this information alongside other measures to guide federal aid funding distribution, Hyslop added.
Data, when shared with policymakers and parents, should come with a caveat around participation rates, coupled with solutions. “Even potentially breaking down which students are missing,” Hyslop said. “That’s giving some additional context around what the data means.”
However, because participation rates varied widely even within states, comparison to previous years’ performances may not always be “apples-to-apples,” according to Marion. In this case, some states are using the “fair trend” method, where performance from students in 2021 is compared to the performance of their peers from 2019.
What the data shows
Results from every state the Center for Assessment has analyzed, about 15, using either the fair trend method or other methods, showed significantly slowed academic progress, consistent with previous predictions and data collections.
“I was one of those skeptics [saying], ‘Don’t we already know enough?’” Marion said. “But I think politicians pay attention to test scores. Policymakers do and state education chiefs do, and so I think that this is why [testing] was worth it.”
The results, he said, show students are behind by between a quarter to three-quarters of a year.
“That’s big,” Marion said. “If we had a quarter of a school year gained, we would be throwing parties in the streets.”
The negative impact on student learning is two-to-three times as large as the impact of Hurricane Katrina on students in New Orleans. While Marion was originally cautious about interpreting results this year because of the variables and potential misinterpretations, that’s no longer the case.
“The signal is so damn loud that it almost doesn’t matter if there’s noise, because it’s just like pulling the fire alarm,” Marion said. “These kids really got hosed last year — it’s not anybody’s fault except for the pandemic’s.”
Photo: Learning Liftoff
Read his and other stories at K-12 Dive