The Education Gadfly Weekly

As we approach the fourth birthday of the Every Student Succeeds Act (ESSA), much angst remains around testing and accountability in education. ESSA released some steam from the boiling pot of education politics by turning a number of key decisions over to states and reducing the stakes associated with “high stakes testing.” But those politics still simmer. That’s partly because American children continue to spend lots of time taking standardized tests; because of growing interest in aspects of schools’ mission that go beyond academics, especially social and emotional learning; and because of questions around whether testing and accountability systems are actually helping to make our schools better.

There’s an almost understandable urge then to lurch toward yet another reform strategy. In my opinion, that would be a big mistake. Most states have spent the past decade overhauling their standards, tests, and accountability systems, and finally committing real resources to capacity-building, especially in the form of curriculum implementation. These pieces have only begun to come together in the last year or two, culminating with the release of school ratings as required by ESSA. What’s needed isn’t to spin the wheel of education policy once again, but to show some patience and commitment—and finish what we started.

Let’s start with a short history lesson.

Accountability 1.0

For most of the twentieth century, state policymakers tried to do quality control by regulating various inputs and processes. To fend off nepotism in hiring, they instituted elaborate civil-service-style protocols for certification and licensure of teachers and administrators. To maintain the integrity of the high school diploma, they required students to pass certain courses and earn a specified array of Carnegie units. To make sure dollars were targeted to needy kids, they put in place elaborate accounting rules.

Some of those efforts may have helped, especially in communities sorely afflicted by corruption and cronyism. But they created a whole new set of problems by wrapping schools and educators in miles of red tape. Labor agreements that ran hundreds of pages long made this problem worse. All of this created a culture of compliance, rule-following, and box-checking, rather than a can-do attitude of innovation and continuous improvement.

By the 1980s, some leaders, frustrated at the lack of progress, proposed a new approach. It was what then-Governor Lamar Alexander called an “old fashioned horse trade”: Policymakers would stop trying to micromanage every aspect of our schools; in return, schools would be held to account for their results.

Around the same time, some groups on the left had also grown frustrated with the regulatory approach and embraced accountability as a tool to force schools to pay attention to low-income kids, children of color, and other disadvantaged youngsters. This gave rise to the left-right coalition led by civil rights and business groups that supported testing and accountability in the 1990s and gave birth to the No Child Left Behind Act of 2002.

Rather than focus generically on holding schools accountable for results, these reformers wanted to hold them accountable for equitable results. Closing achievement gaps (and getting low achievers up to minimal standards) became the priority. So what I will call Accountability 1.0—up and through NCLB—reflected that.

The early attempts at results-based accountability now look rudimentary, even primitive. As modeled by Texas and North Carolina, and later mandated by the 1994 Improving America’s Schools Act, students would be tested regularly (at least once in elementary, middle, and high school), and schools would be categorized based on the percentage of students passing the low-level tests. We did not yet have annual testing that would enable us to gauge progress over time, and disaggregation by student subgroup was a Texas-only phenomenon. Nor were there yet many mandatory interventions in failing schools.

Still, the principle had been enshrined in policy. No longer would we view school quality through the prism of inputs and resources alone—degrees earned by teachers, the size of classrooms, the number of books in the library. Instead, we would look at outcomes.

No Child Left Behind took this approach to another level, with its annual assessments in grades three through eight, rules for measuring “adequate yearly progress,” a universal proficiency target of 2014, and a mandatory “cascade of sanctions” for chronically low-performing schools. Its entire design was focused on getting all students to basic levels of literacy and numeracy. Neither the standards in place in most states nor the assessments were very challenging. Nothing in the accountability model incentivized continued progress for kids who had already mastered low-level standards. But studies would eventually show that the approach largely advanced its mission: The performance of the lowest-performing students rose dramatically from the 1990s into and through the 2000s. Those students are now achieving two to three grade levels ahead of their earlier counterparts. This is historic, life-changing progress. (The booming 1990s economy and big spending increases into the 2000s were at least partly responsible for this good news.)

Given our vast education system and limited surveys about what was actually happening in our schools, it was hard to know for sure what caused achievement to rise so dramatically for our lowest-performing students in the 1990s and 2000s. But it certainly gave reformers momentum, as well as confidence that progress was possible.

Accountability 2.0

With the Nation’s Report Card showing students making gains, and with George W. Bush still in the White House, support for accountability remained relatively strong for the first half of the 2000s. But as the 2008 presidential campaign heated up, NCLB and related policies started to attract more criticism and derision. The problems of testing and accountability were becoming all too clear. Among them: many schools were spending hours upon hours on low-level test-prep; districts had started using commercially developed “formative” and “diagnostic” tests, in part to gauge how their kids were likely to do on the state test, which led to a perception of over-testing; and the “one snapshot in time” approach to measuring effectiveness was unfairly punishing schools serving lots of disadvantaged students while tacitly discouraging schools from paying attention to everyone but the “bubble kids” performing near the line demarcating “proficiency” in reading and math. Perhaps most significantly, the low-level standards and tests in place in most states were sending the wrong signal to parents, educators, and taxpayers: that vastly more students were on track for future success than really were. It was the illusion of proficiency.

With many fits and starts, these concerns eventually paved the way for Accountability 2.0. This recalibration included much more demanding academic standards that were aligned to readiness for college and career, conspicuously in the form of the Common Core; much higher-quality and more rigorous assessments, in the form of Smarter Balanced and PARCC and their successors; and much fairer accountability systems with a greater focus on student progress over time, which was first allowed under NCLB waivers, and then ESSA, plus more transparent school ratings, like five-stars or A to F. It also took most of the “tough” out of accountability’s “tough love” approach, by toning down the expectations for what states have to do when faced with chronically low-performing schools.

This transition has taken the better part of a decade; only in 2018-19 did most states have all of these components in place and issue their first school ratings under ESSA. The question going forward is whether today’s strong economy will combine with Accountability 2.0 to get student progress back on track.

The Work Ahead

“Accountability” doesn’t directly cause students to learn more. Only new behaviors in the real world, practices engaged in by teachers and kids in particular, can do that. But we accountability hawks hope that external pressure will nudge schools toward those practices and overcome the inertia of the status quo.

By raising standards and making the state assessments tougher, we hope that teachers will raise their expectations for their students. That means pitching their instruction at a higher level, giving assignments that ask children to stretch, and lengthening the school day or year for kids who need more time to reach the higher standards.

We also hope that curriculum developers will create instructional materials aligned to the new, higher standards, materials that would encourage a more challenging and effective type of classroom instruction. This hope has already come to partial fruition, as publishers both old and new release programs that expert analysts at Ed Reports have found to be aligned to the new standards. Now there’s a ton to do to get these materials to teachers and help them work with their colleagues to master them.

All of this, we hope, will rekindle the progress of our lowest-performing students, but also lead to gains for kids at the middle and the top of the performance distribution. Thus the shift from “no child left behind” to “every student succeeds.”

None of this will be easy or automatic. It will take time and leadership and investment, and we will need to wrestle with difficult challenges and trade-offs. For example, now that we are aiming for much more challenging material, schools are finding that many of their students are achieving well below grade level. How to meet these students where they are while accelerating their progress toward where they need to be is extremely difficult work. It’s contentious, too, as it’s not always clear when to “personalize” or “differentiate” instruction, especially for low-achieving students, or when to teach them grade-level material even when it’s a stretch.

This is the nitty gritty work of improving teaching and learning. Digging in and figuring it out, helping educators share what they are learning and get better at getting better, isn’t particularly sexy. Nor will it scratch the anti-testing itch. But it’s what our kids need most if we’re going to recapture the progress we had been making in reform’s earlier days. Relitigating old fights over accountability would only be a distraction.

Stay the course. Do the work. Help every student succeed.

This article was adapted from a longer essay published this week in the Phi Delta Kappan.

The racial integration or segregation of K–12 schools is again a debate topic in education circles. Today’s controversy has a new twist: casting charter schools as the main antagonist to integration, claiming they resegregate public schools.

This accusation obscures how three factors not controlled by charters interact to shape a school’s racial composition: housing patterns, school district boundaries, and changing school-age demographics—i.e., an increasing number of nonwhite students and fewer white students attending public schools. Additionally, many charters are specifically created to serve students in communities with high proportions of students of color, communities that are often desperate for quality schools.

Analysts at the Urban Institute have produced what Mathematica Policy Research Senior Fellow Brian Gill calls the “first credible ... methodologically strong study” describing how these factors interact. They calculate racial and ethnic enrollment for grades K–12 at nearly every public school from 1998 to 2015, examining how seventeen years of racial and ethnic data change over time when linked to four geographic units with different housing patterns: traditional school districts, counties, cities and towns, and metropolitan areas.

First of all, charter school expansion nationally has not produced higher levels of segregation, though it has effects in some communities. In their words: "Charter school growth … has increased racial and ethnic segregation in schools. [But] this effect is modest—eliminating charter schools … would decrease segregation 5 percent. Moreover … charter schools reduce segregation between districts in the same metropolitan area.”

There are state differences. Charters in Arizona, Florida, Georgia, New Jersey, and Oregon have little or no effect on segregation. Significant segregation effects occur in Louisiana, New Mexico, North Carolina, and Oklahoma.

This discussion often omits two innovations fostering school integration: common enrollment systems and diverse-by-design charter schools. Common enrollment systems, also called unified or universal enrollment, create a uniform registration procedure for both district and charter schools of choice. Schools share a website, application, and timeline for submitting applications and announcing acceptances.

They also include information on school performance and programs. Those operating these systems provide families with assistance to navigate the process.

A computer process following predetermined rules (in computer-speak, an algorithm) matches schools and students. This algorithm can include criteria that varies the school’s racial, ethnic, and socioeconomic makeup.

These systems are also different from each other. For example, the number of schools parents rank on applications range from five in Denver to twelve in the District of Columbia. New Orleans provides applications in different languages. Camden and Indianapolis do not. Some systems (Denver, for example) are operated by the school district. Others are operated by local nonprofit organizations, as in the District of Columbia.

Another promising charter strategy is diverse by design schools. The recruitment efforts of these charters are designed to reach specific families, students, and geographic areas, giving preference (or weighting) to factors that attract a mix of students.

The Diverse Charter School Coalition represents over 175 schools enrolling 50,000 students. Its members affirm that “charter schools can and should contribute to solving the historic challenge of integrating our public-school system.” Its School Launch Program recruits and prepares school leaders for new diverse-by-design schools. The Century Foundation found that one in five charter schools nationally demonstrated some consideration of diversity in its school model, and one in four had medium or high enrollment diversity.

An example of this diverse-by-design school is Rhode Island’s Blackstone Valley Charter School Prep, opened in 2009 after Democratic Mayor Daniel McKee of Cumberland, now Rhode Island’s lieutenant governor, led a regional coalition to create the school. It operates three elementary schools, two middle schools, and one high school, and it is home to 2,100 students and 300 teachers. This regional model recruits students from a broad geographic area: the more affluent communities of Cumberland and Lincoln and less affluent communities of Central Falls and Pawtucket.

Charter schools are not the main antagonist in today’s school integration/segregation controversy. They are at the mercy of existing housing patterns, school district boundaries, and changing school-age demographics. Innovations like common enrollment and diverse-by-design schools, pioneered by the charter sector, point to new ways to integrate America’s K–12 schools and advance equal educational opportunity.

Editor’s note: This essay was first published by the Washington Examiner.

With school reform seemingly stuck in reverse, helpful hints on the path forward might lie beyond education’s boundaries. To wit, Ash Carter recently sat down with NPR to talk about his new memoir, Inside the Five-Sided Box: Lessons from a Lifetime of Leadership in the Pentagon. Carter, who served as Secretary of Defense during the waning years of the Obama administration, offered a number of fascinating insights into the safeguarding of national security, but none were more interesting to me than the “CRIKT” acronym—an ostensible evolution of W’s “Axis of Evil”—used to identify our nation’s biggest security threats.

Here’s how Carter describes it in his book:

As SecDef, I began talking about the fact that the United States faced no fewer than five major strategic challenges: China, Russia, Iran, North Korea, and the threat of terrorism…

The CRIKT formula sent some clear signals that helped people at DOD and beyond make sense of other strategic and organizational initiatives I was championing. For example, it made clear that my advocacy of more high-technology systems, in areas ranging from drone warfare to cybersecurity, was driven by concrete strategic threats, not just a vague sense that the passage of time required technological improvements.

Carter’s plainspokenness got me thinking whether there’s a parallel here in education. After all, there’s plenty of head nodding to go around when education talk turns to gaps in knowledge and achievement or high expectations and higher order thinking, but we rarely talk about education’s “adversaries” in any concrete sort of way.

In the spirit of CRIKT, I see at least four major strategic challenges facing education improvement in this country, the acronym for which is a bit tongue-in-cheeky: CUSS. Granted, some of our challenges are more technical in nature, like the nuts and bolts of recruiting good teachers or implementing strong instructional materials. However, all of this is made harder than it should be by the CUSS groups.

1. Curriculum publishers. While there’s a lot of promise in the current effort to improve instructional practice, one cannot ignore the outsized role played by legacy textbook firms in determining how this effort will play out. These publishers have sales reps from coast to coast—often former superintendents with personal relationships with the schools and districts they are marketing to. The result is a curriculum adoption process that focuses more on wining and dining prospective clients and less on the procurement of an effective product. As my colleague Mike Petrilli has written, only 10–15 percent of districts today use the good stuff, according to an Education Week analysis of market share against various English language arts and math curricula reviewed by EdReports. And increasing that number substantially is far from given. More research is needed, but the evidence we do have suggests that curriculum makes a big difference and is more cost effective than other interventions. Yet we remain oddly reticent about making long-term commitments to helping teachers access and utilize high-quality curricula.

2. Unions. If there’s a Goliath to be slayed, the unions and their sympathizers represent the single greatest force of resistance and repeal. The need for a political counterweight to their might and muscle is more pronounced today than it has ever been. Consider Exhibit A in the Democratic primary for president, Exhibit B in the national wave of labor unrest, or Exhibit C in Providence’s “thick contract.” The Washington Post’s David Von Drehle recently observed, “We’ll know labor is headed in the right direction when the National Education Association becomes an engine of school reform.”

3. School boards. For better or for worse, a laissez-faire federal posture means that the real gatekeepers to school improvement lie with these locally elected bodies. Never mind that too many are notorious for wandering where they don’t belong. Author Dylan Wiliam recently wrote, “For now, it is school boards that are going to be the most important drivers of change (though I wish that weren’t the case!).” I have yet to come around to his point of view, but it behooves us to be clear-eyed about their ability to make or break just about anything they touch.

4. Schools of education. Without being flippant, higher education often makes K–12 look like it’s got its act together. Nowhere does this ring truer than within our nation’s schools of education, where denizens largely have their feet planted in midair as they proffer alternate forms of science, most egregiously on the topic of reading. Many among our ranks have waxed eloquently about the need to reform the education system at its source, but the mechanism for getting our schools of education on board remains ever elusive.

There are arguably additional challenges, but regardless of whether you’re a booster of social and emotional learning or school choice, one or more of these four will likely pose a threat. The question is whether we learn from previous mistakes made with each of these hazards, and how to adequately prepare for them as we press forward on today’s initiatives.

On national security, Carter argues that each challenge represented by CRIKT is different, and that recognizing these differences is the first step in addressing the problem. He says that if you go to a military briefing in China or Russia, the topic is always the United States. If you go to one at the Pentagon or the White House, it’s invariably about CRIKT. As Carter succinctly puts it, “We have to be concerned about all of them, because they are all focused on us.” Similarly, we should be under no illusion about CUSS and their focus on preserving the status quo and defeating reformers. Until we achieve clarity on that, you can expect the wheels on school reform to continue spinning backwards.

It’s a bit of an education cliché to say “every teacher is a literacy teacher.” Since background knowledge is a fundamental building block of language proficiency, it’s technically true: A teacher in any subject can’t help but be a literacy teacher, even if the effects are diffuse. But a new report from the Education Endowment Foundation (EEF), based in London, England, provides a good road map for secondary school subject area teachers who want to maximize their contributions to student literacy—and reap the benefits of enhanced student literacy in their subject-matter classrooms.

The brief makes strong case for what the authors call “disciplinary literacy,” an approach for improving literacy across a school’s curriculum, starting with the recognition that “literacy skills are both general and subject specific.” The authors present seven principles or recommendations for literacy instruction “grounded in the specifics of each subject.” These include prioritizing disciplinary literacy across the curriculum; combining writing instruction with reading in every subject; developing students’ ability to read complex academic texts; and providing targeted vocabulary instruction in every subject.

If all of this sounds obvious or anodyne, consider that “literacy” tends to be viewed as the exclusive concern of English teachers if not elementary school teachers long before students showed up in secondary school classrooms. The authors, EEF’s Alex Quigley and Robbie Coleman, are writing for British educators, but this view of literacy as not my job is almost certainly as common among subject-specific middle and high school teachers in the U.S. as the U.K.

“The emphasis on disciplinary literacy makes clear that every teacher communicates their subject through academic language, and that reading, writing, speaking, and listening are at the heart of knowing and doing Science, Art, History, and every other subject in secondary school,” the authors note. This begins with an assessment of the literacy requirements of all academic subjects and teachers asking what is unique about their subject discipline in terms of reading, writing, speaking, and listening; how people in their respective fields use language; and taking an inventory of words and phrases used “typically or uniquely” in each discipline. The word “factor,” for example, may be used typically in math class, but it is not unique to mathematics. A student may be asked to “factor” a number in math class then discuss the “factors” that led to the U.S. Civil War in history.

Here the authors invoke Isabel Beck’s helpful and clarifying “tiers of vocabulary.” So-called “Tier 1” words are the simplest, the kinds of words children often come to their first days of school already commanding—desk, ball, baby, etc. “Tier 3” words tend to be discipline-specific terms like “photosynthesis” or “isotope” that are seldom used outside of particular fields of study. The richness of language tends to reside in high-frequency “Tier 2” words that may occur across disciplines, but that take on different meanings in different contexts. “It is easy to see how confusion for students can occur,” the authors write, in mathematical words like value, prime, area, mean, fraction, and improper, which mean entirely different things in math class and everywhere else. Getting secondary school teachers to think carefully about subject-specific vocabulary and the potential to confuse students is by itself a worthwhile exercise.

Elsewhere the authors dwell on reading strategies a bit more than skill-and-strategies skeptics might wish, but at least take care to note that “some authors have argued that it may be possible to teach reading strategies quickly and then move on.” There is also a case study on “reciprocal reading,” with four students assigned specific roles to muddle through a complex text, which seems cumbersome, unserious, and beneath the dignity of high school–aged students. But in its broad strokes, the report offers a rich array of ideas, many of which American secondary teachers might find useful and effective (the report has me reflecting on the vocabulary demands on my students in a high school civics class that I teach). At the very least, the report provides actionable context and practical guidance for literacy instruction in subject-matter classrooms, which promise to pay dividends in those disciplines. As the authors note cheekily, “Secondary school teachers should ask not what they can do for literacy, but what literacy can do for them.”

SOURCE: Alex Quigley and Robbie Coleman, “Improving Literacy In Secondary Schools,” Education Endowment Foundation (July 2019).

Flipped classroom in K–12 and higher education have been popular for years. Unlike the traditional model in which lectures in class are followed by homework exercises, students in the flipped variety accomplish these in reverse, watching videos of lectures or working through other tutorials at home, and then exercising their new skills and knowledge in the classroom, where the teacher can help them if they run into problems, answer questions, etc. It’s an intriguing idea, but there’s a dearth of research on how the model impacts student learning. A recent study published by Brown University’s Annenberg Institute offers valuable new evidence.

Researchers randomly assigned 1,328 students enrolled at West Point to different sections of two courses: introduction to calculus and introduction to economics. Eighty instructors taught at least one section in each type of classroom—flipped and traditional—to ensure that it wasn’t the instructor that was driving any differences.

The potential impact of the flipped classroom model they studied was limited in two important ways. First, the flipped classroom was only implemented for one unit in the course, after which there was a unit quiz, but this was just a small part of the course. Second—and this serves as an interesting finding, as well—it’s clear that some of the students in the flipped sections were not taking advantage of the program’s resources. A fifth of the calculus students and more than a fourth of the econ students in the flipped sections never watched a single video, even though watching the lectures away from class was the whole point of the model.

There are two main findings about the impacts of flipped classrooms. First, the results of the unit quiz demonstrated a strong positive effect of the flipped model in calculus, but a null impact in economics. Students in the flipped calculus course scored about a third of a standard deviation higher on the mid-term quiz relative to students in the traditional classrooms. But this didn’t apply to every subgroup: Female, black, and Hispanic students and students with lower baseline performance in math saw no improvement under the flipped model. This means that the flipped classroom actually increased achievement gaps.

The second main finding was that there wasn’t a difference in scores on the final exam between students in the two classroom types. The researchers used the final exam grades that corresponded to the unit that was flipped in the treatment group, so potentially there would be more effect than on the final exam as a whole. Still, since only one unit, corresponding to at most a few days of instruction, was flipped for the treatment group, it isn’t surprising that the effects washed out.

All in all, the results are promising for the flipped classroom model, with the important caveat that performance only improved for the student groups that were already the highest performing. Future studies should dig into why these effects differ across student groups, along with the question of how we can get all students to take advantage of the available course resources, even those without the internal motivation to do so.

SOURCE: Elizabeth Setren et al., “Effects of the Flipped Classroom: Evidence from a Randomized Trial,” retrieved from the Annenberg Institute at Brown University (August 2019).

On this week’s podcast, Mike Petrilli and David Griffith talk with Robert Pondiscio about his new book on Success Academy. On the Research Minute, Amber Northern examines whether providing more information to college admissions officers boosts the acceptance odds for low-income students.

Amber's Research Minute

Michael N. Bastedo et al., “Contextualizing the SAT: Experimental Evidence on College Admission Recommendations for Low-SES Applicants,” Education Policy (September 2019).