The Accountability Challenge
American elementary and secondary education remains seriously unwell. That's clear to anyone paying attention. Dismal student achievement has grown far worse in the wake of the Covid-19 pandemic. Learning gaps are wider than ever. Too many graduates are ready for neither college nor the workplace. Chronic absenteeism has become a plague. And American schools are regressing rather than recovering on most of these fronts. But there is a deeper cause for worry underlying these problems: The country's multi-decade commitment to results-based accountability for schools and students has badly eroded and may not be recoverable.
The proposition that schools should be "held to account" for their pupils' learning (or lack thereof) was the bedrock of education reform across the United States for several decades. Almost as fundamental, if less widespread, was the proposition that individual students should not move on to the next level, be it fourth grade, advanced algebra, or college, unless they could demonstrate that they had met the standards of the previous level.
Along with mounting attention to school choice in its many permutations, the achievement-boosting, school-changing strategy erected on accountability (commonly termed "standards-based reform") drove passage of state and federal laws — peaking with 2002's No Child Left Behind Act (NCLB) — as well as billions of public and private dollars, the earnest work of hundreds of would-be education fixers (myself included), the production of innumerable reports and books, and the administration of tens of millions of standardized tests. And it made a difference. Not enough, to be sure, but bona fide achievement gains were made during the last decade of the 20th century and first decade of the 21st. They were mostly confined to math (though reading showed modest bumps as well), but were consequential for many poor and minority students — notably among those at the lower end of the score distribution. Which is to say, the floor rose under those who most needed it.
In no way were such gains sufficient. They didn't touch all children and tended to fade in high school. Yet they were worthy of sustaining and augmenting. For the most part, that didn't happen: Scores were flattening or sagging well before the pandemic. One reason was the gradual weakening of accountability, which has since approached collapse.
THE ROOTS OF ACCOUNTABILITY
The logic undergirding an accountability-centered theory of education reform can be traced to a blockbuster federal study released over the Fourth of July weekend in 1966 (timing carefully chosen by Lyndon Johnson's White House, it was said, to minimize attention to unwelcome findings) and written by the celebrated sociologist James Coleman. In its smaller way, it resembled the work of Copernicus or Galileo, challenging the long-held belief among educators and public officials (President Johnson included) that if you want better outcomes from schools, you add to their inputs: shovel in more money, require more credentials for teachers, shrink class sizes, buy better textbooks, and so on.
Not necessarily so, said Coleman. His voluminous data showed that this assumption was unreliable, that the relationship between school inputs and outcomes was shaky at best, and that other forces (family circumstances, peer group, and more) caused greater differences in pupil achievement than anything done by schools.
Though this conclusion was fiercely resisted by educators, and elements of it remain contentious today, by and large it has held up for almost six decades. Its implications for education policy were profound: If school outcomes — student achievement in its various forms — are unsatisfactory, don't just pump in resources and assume the best; focus instead on the outcomes you seek, orchestrate the work of schools to produce them insofar as possible, and hold teachers and principals to account for how well they succeed.
As tripods should, the "accountability tripod" had three legs: establish academic standards that spell out what kids should learn at each stage of their compulsory education; devise tests or other measures to track how well they are or aren't learning it; and reward schools (and those who work in them) that produce the desired results while embarrassing, punishing, or intervening in those that don't.
Along with school choice, this strategy characterized most American school-reform efforts in the aftermath of 1983's much-cited A Nation at Risk report, which angrily displayed the inadequacies of what schoolchildren were then learning in the K-12 years. Coleman had said to focus on the outcomes you want; A Nation at Risk said our outcomes were sorely lacking. And many governors, especially in the South, had already figured out that transforming their stumbling old agrarian economies meant that they had to imbue their populations with more skills.
Implicit in the "tripod" approach was the assumption that educators knew what to do to turn around slipshod schools and boost student learning but didn't necessarily have the incentives — or often the autonomy — to do what was required. Simply stated, they possessed the skill but lacked the will, except when they had both but were stymied by inertia, regulation, and politics — the latter two usually involving adult interests taking precedence over those of children.
Accordingly, reformers sought to boost achievement by creating incentives and loosening constraints. Or, at least, that was the original idea.
Tennessee's Republican governor Lamar Alexander — who went on to serve as U.S. education secretary and an influential senator — characterized the freedom-for-results approach in 1986 when, summarizing a major report by the National Governors Association, he said:
The governors are ready for some old-fashioned horse-trading. We'll regulate less, if schools and school districts will produce better results....[T]he governors want to help establish clear goals and better report cards, ways to measure what students know and can do. Then, we're ready to give up a lot of state regulatory control — even to fight for changes in the law to make that happen — if schools and school districts will be accountable for the results.
That loosening of control in return for better results is essentially what the then-new charter-school movement did while also advancing school choice. But system-oriented reformers — and federal and state policymakers — tended, as government officials most often do, to make more rules and tighten the screws. Thus the movement for accountability pushed in two opposing directions at once.
The temptation to tighten further would soon follow the Charlottesville "education summit" to which George H. W. Bush — who had vowed to be an "education president" — summoned the nation's governors in September 1989. Hard as it is to imagine today, 49 of them turned up, Arkansas's Bill Clinton prominent among them.
Their product was a set of wildly ambitious "national education goals" that they declared the country should reach by century's end. The third of those goals stated that by the year 2000, "American students will leave grades four, eight, and 12 having demonstrated competency in challenging subject matter including English, mathematics, science, history, and geography." The fourth vowed that "U.S. students will be first in the world in science and mathematics achievement."
These were grand aspirations. But what force might bring them closer to fulfilment? State-led reforms were obviously needed. Yet if this was indeed to be a national effort — promoted, if not led, from the Oval Office — what could Uncle Sam do to help?
Exploiting the then-recently enhanced authority of the National Assessment of Educational Progress (NAEP) to set performance benchmarks on its tests and report state-level outcomes was one way. The presidential "bully pulpit" was another. But what about something with more oomph that would lead to change across the entire at-risk nation?
GOALS AND RULES
The idea came on gradually, then accelerated. In April 1991, Bush and the ubiquitous Alexander, now secretary of education, unveiled their four-part "America 2000 strategy" for nudging the country toward those aggressive goals via innovation, school "reinvention," more choice, and "voluntary national tests" — plus a lot of cheerleading and pulpit-pounding.
Congress mostly yawned in response, but perked up when Clinton (a Democrat) became president and proposed enshrining the goals in federal law. This happened in 1994 with passage of the Goals 2000: Educate America Act. Clinton's education secretary (and former South Carolina governor) Dick Riley would describe that measure in these gentle terms:
Instead of the traditional federal regulatory approach, it was a new model of partnership with states and local school districts that encouraged innovation, promoted flexibility and cut red tape....Public education would remain primarily a state and local responsibility, but this law established that it is indeed a national priority. It's important to note that the Goals 2000: Educate America Act was administered without writing a single new regulation.
This formulation was a bit disingenuous because that same year — weeks before Newt Gingrich's "Republican Revolution" won the GOP both houses for the first time since 1952 — Congress also passed the Improving America's Schools Act, which reauthorized LBJ's Elementary and Secondary Education Act with more strings and conditions than ever before, this time affecting the billions of federal school-aid dollars administered through what's known as "Title I." As the bill summary stated, this law
Requires any State desiring to receive a grant...to submit State plans that...describe challenging standards for all children that will be used by the State, its [local education agencies], and its schools to carry out this Act, including: (1) challenging academic content standards; and (2) challenging student performance standards....Directs the Secretary of Education to establish a process for peer review and Secretarial approval.
The screws were tightening — and this, at the outset, had a positive effect. The education arena of the 1990s buzzed as states, mostly for the first time, set academic standards for their public schools, created tests aligned with those standards, in some cases created school accountability plans attached to test results, and erected hurdles for students, such as tests they must pass in order to obtain high-school diplomas. But with substantial federal dollars now at stake, these mandatory "state plans" received ever-closer federal oversight.
As the millennium began, Uncle Sam bore down harder still. George W. Bush brought with him to Washington the conviction that the entire country should emulate Texas's impressive record of boosting achievement and narrowing racial gaps via its accountability tripod. He enlisted bipartisan support — notably Massachusetts Democrat Ted Kennedy — and passed NCLB at the end of 2001.
The centerpiece of that much-discussed statute — a near repudiation of the "horse trade" approach that Alexander had outlined — was a steep staircase of achievement targets for individual schools tied to a cascade of interventions in those that failed to scale the heights. It all hinged on regular testing of reading and math, followed by "disaggregation" of results such that no school could be considered good if any of its subgroups were performing poorly. While states would set their own standards and select their own tests, the federal Department of Education had to approve the latter, and the NAEP would show whether their standards and test "cut scores" were truly rigorous.
Thus by 2002, "school accountability" had become a sort of hierarchy in which faltering performance by schools (with the prominent exception of charter schools) would be dealt with by districts, districts by states, and states, while not directly intervened in by Washington, would have their performance audited and publicized via federal tests, and would need the Education Department's approval of their own tests and plans in order to continue receiving significant sums of federal dollars.
Meanwhile, at the student level, several states adopted uniform end-of-course exams and all states had to report to parents their children's performance on the tests mandated by NCLB. Some jurisdictions adopted "reading guarantees" such that children could not proceed from third to fourth grade until they passed a statewide assessment. "Kindergarten readiness" gauges were established in many places, and more states — 27 at the peak — imposed graduation tests.
The best-known and arguably most successful example of a student accountability program is Massachusetts's Comprehensive Assessment System (MCAS), launched in the '90s as part of a wide-ranging reform effort that has so far been maintained (with adjustments) under governors of both parties and a revolving cast of legislative and education leaders. Such perseverance with an education strategy of any sort is a rarity in a land where elections usually lead to policy shifts and local superintendents are soon replaced by leaders with different agendas.
MCAS is not very demanding. It's basically an eighth-grade level test — subject to various work-arounds, exemptions, and opportunities for retakes — of English, math, and science that high-school students take starting in 10th grade. But it has lasted almost three decades, and most observers credit it for the "Massachusetts miracle," as it was termed, particularly in the late 1990s and early 2000s when Bay State NAEP scores rose dramatically, surpassing those of all other states and comparing favorably to hard-charging Asian nations on international comparisons such as the Organization for Economic Cooperation and Development's Program for International Student Assessment, or PISA, exams.
THE BEGINNING OF THE END
Perhaps predictably, this shift toward accountability — and the tests and judgments it depended on — led to backlash, as government-enforced behaviorism always does. Too many schools were labeled "in need of improvement" under federal criteria, even if they were satisfactorily educating most of their pupils — and more than a few of those schools were located in leafy suburbs where property values are commonly associated with a reputation for "great public education." Stressing reading and math scores felt like it narrowed the curriculum, even as days spent prepping students for annual tests irritated teachers and cramped their style. The focus on getting kids over the "proficient" bar led to neglect of those already well over it. It also tended to be unfair to low-income schools that were better at boosting their pupils' weak skills than at boosting them over the bar.
Teachers didn't like it, particularly when Barack Obama's Race to the Top program — part of 2009's big economic stimulus passed to deal with the Great Recession — caused states to judge individual instructors based on their pupils' scores (even though experts cautioned that the then-current testing system was a valid evaluation for just a fraction of the teaching workforce). The attempt was enough to cause teachers and their unions to rail against both accountability and testing, even as rekindled attention to "equity" focused more spotlights on the achievement gaps that those cursed tests revealed. More than a few progressives favored shooting the messenger.
Race to the Top also infuriated conservatives, many of whom didn't care for Obama in the first place. The fact that the program effectively coerced states into adopting the "Common Core" standards for reading and math didn't help, either. The standards themselves had merit, being meatier and more rigorous than most state standards, and they had emerged from a state-driven development process. Yet they smacked of a national curriculum imposed by Uncle Sam, meaning hostility from the right was all but inevitable.
Thus arose a perfect storm over results-based accountability and the academic standards and tests it depended on. In addition to teachers and real-estate agents, many parents (surely egged on by teachers) disliked the whole regimen. Kids didn't like it. District bureaucrats didn't like how it threatened their power base and disrupted their routines. All this animosity would intensify further during the Covid-19 pandemic, when just getting kids (and teachers) into schools felt like achievement enough.
Long before that, however, the Department of Education began issuing waivers and allowing modifications that eased NCLB's rigid framework while pumping out billions of dollars for "school improvement" programs that yielded meager results. It turns out to be extremely difficult to turn around a low-performing school, save when the interventions are so drastic — changing the staff, replacing the curriculum, transforming it into a charter school, and so forth — that few districts or states have the guts to follow through. Instead, they typically opted for the mildest permissible interventions, such as telling the current school team to draw up an "improvement plan."
Mildness got further license from Washington in 2015 when Congress, in response to the various rigidities in and protests against the 13-year-old NCLB, relaxed the federal rules and consequences for low-performing schools, leaving states to determine in large part what (if anything) to do about them and their districts. One can reasonably interpret this as an overdue return to the 10th Amendment and a corrective to federal overreach. But the practical effect was that states, and blue states in particular, could revert to laissez-faire.
We also learned along the way that the tripod needed a fourth leg, for it turned out that many educators lacked the requisite know-how to transform their schools and amp up their students' learning. They had never needed to acquire such skills, or their supervisors didn't care about them, or their education-school professors had failed to supply them — maybe didn't even believe in them.
A more comprehensive reform strategy would have incorporated elements of what education-policy types term "capacity building" for teachers and administrators alike, as well as school-board members and others who set policy for school systems. It would pay greater attention to curricula and other "inside the classroom" elements. And — as in Alexander's "horse trade" — it would lower the barriers to change enshrined in long-standing regulations, certification rules, collective-bargaining agreements, and the like.
But capacity building tends to cost money, takes a long time, and smacks of fiddling with inputs rather than focusing single-mindedly on outcomes. So not much of it happened — which only added to educators' frustration.
The turn against accountability received an extra push during the Donald Trump/Betsy DeVos era in Washington, which emphasized school choice over standards-based reforms and further weakened the spirit of bipartisanship that had produced such measures. The arrival of Joe Biden and Secretary Miguel Cardona helped not at all, for they've shown almost zero interest in student achievement while showering attention on every form of equity and shoveling billions of "Covid-19 recovery dollars" into public schools with little obligation to spend it on recovering student achievement.
The Biden team reflects an old, pre-Coleman, LBJ-style approach to education (always the preferred approach of teachers' unions), which is to go heavy on resources and equality. Meanwhile, most conservatives seem to have pledged their troth to school choice and marketplace forces. Hence there's very little constituency, at least among policy leaders, for standards or testing, even if school accountability in the abstract continues to do well in public-opinion polls.
THE END OF ACCOUNTABILITY
Evidence of this waning attention abounds. Today, only eight states have any sort of high-school graduation test. The head of FairTest gloats over this victory for anti-testers, declaring that such requirements "harm thousands of young people who either drop out after failing an exit exam or are forced out of school without high school diplomas despite completing all their classroom work."
In the Bay State, the MCAS is now in jeopardy, as anti-testers and equity hawks have joined forces with the teachers' union to press for its abolition, and more lawmakers have come to agree. As state senator Joanne Comerford put it, "a single test shouldn't determine if a student graduates from high school....This is a test that disproportionately fails our most vulnerable and at-risk students....We need to bring an end to punitive high-stakes testing."
Things aren't much different in the Midwest, where the same argument convinced Ohio legislators in June to defang the Buckeye State's "third grade reading guarantee" — the requirement that students pass a reading test before being promoted to fourth grade. Going forward, they won't be held back unless their parents consent, which isn't apt to happen very often. In the same omnibus bill, the state's deep-red General Assembly enhanced school choice on several fronts.
Meanwhile, Georgia didn't abolish its four statewide end-of-course exams, but they don't make much difference anymore: Peach State lawmakers recently voted that districts may count those scores for as little as 10% of a student's grade in the relevant classes, which means credit toward graduation can mount faster, even when students flunk state tests.
It's important to understand that neither single-course state tests nor MCAS-style graduation requirements nor "reading guarantees" (still found in several states, including Mississippi with its own recent "reading miracle") are cruel, insensitive, or rigid. Tough, yes, but invariably accompanied by exceptions, second chances, and remedial arrangements for kids who for some reason just can't pass the regular test on schedule. The point of such measures isn't to punish. Amply supported by studies that demonstrate their efficacy, their purpose is to ensure that students moving on to the next stage of their educations — or their lives — have learned the essentials that they will need to succeed there. If schools aren't tough about that now, these students will run into tougher problems later. That's tough love.
The pandemic brought "accountability holidays" of every sort — even the federally mandated annual testing was suspended in 2020, and the 2021 version allowed for wide variation in how it was administered. States relaxed or skipped the identification of low-performing schools. Many schools stopped grading or failing students, or ordered teachers not to score anything below 50% — all this on top of the grade inflation that was its own epidemic long before the coronavirus kicked up, as was the suspension of entrance-test scores by ever more student-hungry colleges.
Many states, meanwhile, had already made use of their post-2015 federal freedom to ease up on school accountability, particularly any sort of intervention in failing schools. (Some, though, ramped up the "embarrassment" version by emulating Florida's move to assign old-fashioned letter grades to schools based on their performance, the theory being that parents seeing D or F grades applied by the state to their children's schools might be motivated to do something about the situation. Educators of course hate this practice.)
I happened to be on Maryland's state board of education as the Old Line State was devising its plan (for submission to Washington) in response to that new flexibility. Every time board members suggested — even mentioned in a public meeting — a high-traction school intervention, such as converting a chronically low-performing school into a charter school, the ever-vigilant teachers' union trekked to the overwhelmingly blue-hued and union-friendly legislature and got such possibilities scotched. They also persuaded lawmakers to expand the board with more parent and teacher representatives. When I told Senator Alexander that the federal law he had nursed into passage was having such undesirable effects in my state, he smiled and suggested that perhaps we should elect different legislators.
The country may be headed toward separate education-policy strategies in red and blue states, but traditional results-driven accountability doesn't loom large on either agenda — choice drives the former, equity and money the latter. Nor are learning outcomes at the top of the agenda for anyone who counts in today's Washington, or even increasingly among the advocates and advocacy groups that pressed it forward for decades.
An illustrative case is a recently published report titled "The Path Forward for School Accountability" from the generally estimable non-profit National Center for the Improvement of Educational Assessment. Its 18 pages are long on "stakeholder" engagement, "principled design processes," "support systems," "improvement planning," and "comprehensive" approaches. Nothing in it is objectionable, and much of it is worth taking seriously. But one searches in vain for consequences, for interventions, for what to do to or about a school that may lack the will even when it possesses the skill.
All of this is consonant with a phase in our education history when we find ourselves replacing academic achievement, gains, and gap-closings with whole-childism, "multiple measures" of performance, waivers from testing, and holidays from consequences while focusing on "supporting" schools and "building their capacity" without also holding them to account for their performance. Much love, no tough.
Yes, we probably went overboard on one-size-fits-all goals and metrics, command-and-control remedies, and a big-government approach to the vast, sprawling, variegated world of K-12 education. Yes, we paid too little attention to that loosely coupled system's capacity to change. But now we're overcompensating. Like a golfer who overcorrects his swing, we're missing the green — and likely to keep doing so.
One might think the acute learning losses wrought by the pandemic over the past several years and our inability thus far to correct them might jar policy leaders back to school and student outcomes as the foremost goal. But that's not happening; instead, we continue to soften.
Driving constructive change in a free society is not easy — make too many people do things they don't want to do, and you'll probably get voted out of office. But it's not impossible. Restaurants get shut down by the health department if they have vermin. Pilots and bus drivers get grounded if they can't pass the licensing test. You can't cross bridges that fail safety inspections or inhabit a house that doesn't meet the fire code.
We can be tough when health and safety are at stake. Why would we keep a school open when its students aren't learning? Why would we "license" a student for fourth grade or graduation when he cannot pass the test?
We need a rebirth of accountability in American K-12 education, while also still taking "capacity" seriously. Skills are necessary to that end, for sure. But when the will isn't there, consequences must follow.