The earliest wave of plans submitted by states to carry out the Every Student Succeeds Act are raising stark concerns among advocates that the much-vaunted “guardrails” meant to protect the most vulnerable students might not be as sturdy as they seem.
“You can see states taking ownership of the flexibility that they have in the new law, and you can see that they’re really trying to drive toward more equitable systems of education,” said Phillip Lovell, vice president of policy development and government relations at the Alliance for Excellent Education.
But, he cautioned, “details matter, and those details don’t always add up to the equity vision that most states have.”
Back in 2015, the grand compromise of ESSA was to grant states more authority over K-12 education than they’d had in nearly two decades. On the other side of that bargain were provisions designed to safeguard the students who have traditionally struggled in America’s public schools, including children of color, English-language learners, students from low-income households, and those with disabilities.
Under ESSA, states must provide “comprehensive support” to the bottom 5 percent of schools in the state, as judged by new accountability systems, and to high schools where fewer than two-thirds of students graduate. It’s the third “guardrail,” targeted interventions at schools that are failing one or more groups of historically underserved students, that’s troubling advocates.
The law leaves it up to states to define “persistently underperforming” subgroups, and that has left some big holes, observers say, in how some states are holding schools accountable for their at-risk students.
“The schools that might be doing OK overall, but might not be serving a group of students well” are what the Leadership Conference on Civil and Human Rights is focusing on, said Liz King, senior policy analyst and director of education policy.
Rules issued by the Obama administration to beef up ESSA’s accountability provisions, often to make sure the achievement levels of disadvantaged students could not be masked, were first put on pause as part of a broader Trump administration freeze on regulations. Congress then overturned them in early March through the Congressional Review Act.
The first round of state ESSA implementation plans was due to the U.S. Department of Education April 3. Some states’ submissions have been delayed because Education Secretary Betsy DeVos issued a new template March 13, and the underlying law mandates that governors have 30 days to review the final plan before it’s sent to federal officials.
Roughly 15 states, including Colorado, Louisiana, and Massachusetts, as well as the District of Columbia, have either filed their plan with the department or will do so as part of the first deadline. Advocates said they did not want to say which state plans they were spotting early accountability weaknesses in because many are not final yet or they hadn’t yet had a chance to read all of them in depth.
Some states are defining a group as consistently underperforming only if it’s doing as poorly on a given measure as the bottom 5 percent of students statewide, said Natasha Ushomirsky, director of K-12 policy development at Education Trust.
“Obviously, that communicates a pretty low expectation across the board,” she said.
Others are identifying consistently underperforming subgroups only as they compare to the same subgroup statewide, which “sets a different expectation for every student group,” she said.
Both of those scenarios could leave schools with huge achievement gaps without targeted interventions, so long as the lower-performing group was outperforming the worst schools in the state or their peers statewide.
Still others, Lovell said, are effectively requiring a group to be underperforming in every measure before it will be targeted for intervention.
“Whether that’s an overt requirement or a function of the math, we’re concerned with that becoming a reality for kids,” he said.
Then there’s the issue of whether — and how — the performances of student subgroups are counted in a school’s overall rating.
Some states are very clear that a school’s proficiency rating, for instance, includes both school-wide results and the results of each subgroup. Others have essentially set up parallel structures for overall ratings and the measures of subgroups for targeted supports.
“In an ideal world, you want the school to get one rating at the end of the day. If that school is not doing well for one or more groups of students, then you don’t want it to get a high rating overall, because that allows the school to essentially ignore its underperformance,” Ushomirsky said.
Lovell was more blunt: “You shouldn’t be able to get an A while having a graduation rate for African-American kids of 60 percent.”
Which kids take tests — and which get counted
There are still other ways a plan can shield the performance of some student subgroups.
One is how states will enforce the requirement that 95 percent of students in each school, and 95 percent of each subgroup, take the tests required by law. Some state plans prohibit schools with less than 95 percent participation from receiving the highest scores, while others will merely require some sort of action plan from schools that aren’t hitting that mark.
The now-overturned Obama administration accountability rules “put a little bit of the meat on the bones” and gave states suggestions for how they could penalize schools that aren’t reaching the 95 percent requirement in the final accountability score. But now that they’ve been overturned, it’s up to states to pay attention, said Candace Cortiella, director of The Advocacy Institute, a group that advocates for students with disabilities.
“We’ll have to see over the coming years how the participation rate of kids with disabilities in state assessments fares. It will be a key indicator of what’s happening. This [students with disabilities] is the first group of kids that schools are kind of happy to find ways to keep them out of the testing pool,” she said.
Another is how the state sets the minimum number of students in a particular group to be counted for accountability or data reporting purposes, often referred to as the “n size.”
If a state selects an “n size” of 15, for instance, a school with 14 special education students wouldn’t be held accountable for their performance.
The law doesn’t specify how big or small the n size must be. The defunct Obama accountability rules required states to explain why they chose a number larger than 30 and to show how many schools might be excluded from accountability for different subgroups of students, based on the state’s proposed n size.
Several states did include that information in early-round drafts written before the rule was overturned, and the data were stark, Cortiella said.
The Every Student Succeeds Act: 5 Things You Need to Know
A draft of Michigan’s plan, for instance, said that only 26.4 percent of schools would meet their proposed n size of 30 for African-American students and 36.4 percent for students with disabilities.
“There comes a point at which, between the n size and the participation requirement, it really almost doesn’t matter what the accountability system looks like,” Cortiella said. “If two-thirds, or three-quarters, of kids in a subgroup are allowed to escape that system, really what difference does it make how good it is?”
Another problem, advocates said, is states’ decision to use “super-subgroups” that combine the results of multiple high-needs groups; for example, counting all English-language learners, special education students, and students from low-income homes in one group. The use of super-subgroups was banned in the Obama accountability rules.
That, advocates said, obscures consequential differences among the students thrown together into one accountability pot.
“Those kind of combinations really ignore meaningful distinctions in the needs, as well as civil rights protections, that are afforded to those students,” Ushomirsky said.
There’s also a problem in some places with the big picture — they’ve set different long-term goals for academic performance and high school graduation rates for each subgroup of students.
“There is no excuse for different long-term goals. A state’s objective should never be that they have a discriminatory system, frankly,” King, of the Leadership Conference, said. “In theory and in concept it’s one thing, but on paper you’re telling me that you, as an educator, are comfortable with a scenario that aspires to leave whole groups of kids behind? … That’s just not why educators go to work.”
Some bright spots
Though civil rights advocates have concerns, they’re not universal.
Ushomirsky, for instance, said advocates’ fears that states would skew accountability systems away from academic measures, or use less rigorous measures for their “fifth indicator,” or measure of school quality, haven’t come to pass.
“That’s a definite highlight,” she said.
Most states are using chronic absenteeism as at least one of their measures of school quality, and several have also included measures of college- and career-readiness for high schoolers, such as how many are taking and passing Advanced Placement and International Baccalaureate courses, or career and technical education programs.
Mike Magee, CEO of Chiefs for Change, a reform-minded group of state education leaders, highlighted the use by some of its members — including Louisiana, Massachusetts, New Mexico, and Washington, D.C. — of a summative score for schools, and a commitment to include growth measures while not losing sight of proficiency rates.
Growth is important, but “there are also milestones that students need to reach to achieve their ultimate goals,” he said. “You want that high standard and those high expectations to be reflected in the summative ratings you give to schools.”
Ultimately, the ESSA plans should be living documents that continue to strive toward educational equity, particularly since the law is likely to be in place for 10 years or more, Lovell said.
No Child Left Behind, the federal K-12 law ESSA replaced, lived for some 13 years.
“It’s important that we have a solid start, that this flexibility be used well and that we be willing to not consider what’s done now to be set in stone but we improve over time as we learn more,” he said.