The Education Gadfly Weekly

The national press jumped all over the news last week that the Office for Civil Rights in the Trump Department of Education will be taking a different tack on federal civil rights enforcement than it did under the Obama Administration. As Andy Rotherham wrote, “not surprisingly, with those words—Trump, civil rights, federal—in the same sentence, people are alarmed.” And sure enough, the mainstream media published articles with alarming headlines, like “Education Dept. Says It Will Scale Back Civil Rights Investigations.”

That sure sounds bad; after all, even the local-control crowd will generally acknowledge that there’s a legitimate federal role in ensuring students’ civil rights. Too many of our children and teenagers feel vulnerable today, for no other reason than their race, religion, or sexual orientation, and they deserve our protection.

But what journalists, education reformers, and everyone else should understand is that the Obama Administration turned almost everything into a potential civil rights violation. As I argued at the time, it was federal overreach on steroids. What acting OCR chief Candice Jackson is doing is simply returning OCR to the pre-Obama status quo ante. Whether that’s wise is worth debating. But surely it’s hyperbole to claim, as Jackson’s predecessor did on Friday, that it amounts to “assembly-line justice” and “a stick your head in the sand approach.”

At issue is whether OCR should launch “systemic reviews” when investigating discrimination complaints, or stay focused on the facts of a particular case. As Jackson writes,

OCR will only apply a “systemic” or "class-action” approach where the individual complaint allegations themselves raise systemic or class-wide issues or the investigative team determines a systemic approach is warranted through conversations with the complainant.

In her New York Times article, ace reporter Erica Green points to a California case to explain what’s at stake:

The department received a complaint that a black student at the Lodi Unified School District in California, about an hour south of Sacramento, received harsher punishment than a white student after the two were in a fight.

According to a published settlement agreement, the investigation found that schools with higher percentages of black students established stricter punishment for discipline incidents.

That’s an excellent example. When most of us think about minority students facing discrimination, we picture an event like the one in Lodi—two kids in a fight, one white, one black, and the black student is allegedly given a tougher punishment than his white peer. If true, that’s clearly a civil rights violation, and an issue under the purview of OCR.

Why, then, did the Obama-era OCR feel the need to look beyond this particular event? A cynic might say it was because the facts didn’t point to discrimination. According to the settlement agreement itself, the black student assaulted the white student the next day, and the white student didn’t reciprocate. That explained the differential punishment. So investigators went looking for systemic discrimination, and in their view found it: “Schools with higher percentages of black students established stricter punishment for discipline incidents.”

To researchers this wouldn’t be surprising at all. A brand new study out of the University of Arkansas, like research before it, found that most racial disparities in discipline come from differences between schools, not within them. The scholars write, “Across the state, black students are about 2.4 times as likely to receive exclusionary discipline (conditional on reported infractions and other student characteristics), whereas within school, this same conditional disparity is not statistically significant.” In other words, giving two students in a fight different punishments is not very common, but punishing discipline infractions more severely at high-minority schools is quite common indeed.

But does this amount to systemic racism and discrimination? Common sense would indicate that tougher schools are going to do whatever they can to maintain order and safety and ensure that students can learn. Affluent schools may be able to get away with a more relaxed approach. And if the affluent schools and poor schools are in different school districts, there’s no legal problem.

But what if your district has both high-poverty and low-poverty schools, as is the case for most large districts, either urban or suburban? Here the Obama-era OCR would say that you need to have one district-wide discipline code, and it needs to be enforced consistently—and even then, if it leads to racial disproportionality, that could be grounds for an investigation. In the real world, that means districts have to either practice stricter discipline than educators deem necessary in their low-poverty schools, and/or more lax discipline than educators deem necessary in their high poverty schools. Is that really a good idea, educationally?

Moreover, how can you defend the sort of fishing expedition that OCR embarked upon in Lodi, when the specific complaint was about two students in the same school getting into a fight?

Furthermore, what about the peers of disruptive students? Do they have anyone paying attention to their right to learn while we are parsing the rights of kids who are getting into fights? When districts react to OCR threats by choosing not to enforce their discipline codes in high-poverty, high-minority schools, it’s the well behaving poor kids who suffer “disparate impact.”

***

Whether civil rights rules should allow different schools in the same district to have different approaches to discipline is up for debate. There are trade-offs—pros and cons on either side. But more than anything I hope I have convinced you that using the federal civil rights apparatus to address something as central to schooling as student discipline is…complicated. And what Candice Jackson just did is entirely reasonable, and even kind of normal. Of course, that doesn’t make for good headlines. 

Some smart education reformers just made two thirds of a very dumb mistake. In Charting a New Course: The Case for Freedom, Flexibility & Opportunity Through Charter Schools, Jeanne Allen, Max Eden and others (including Mike McShane, Ben Lindquist, Derrell Bradford and Jay Greene) offer several solid suggestions for state policy makers, such as encouraging more small one-off charters, having more than one authorizer in a given locale, systematically auditing the regulatory burden on charter schools, and giving them latitude to hire the teachers of their choice.

That’s the one third that’s smart and timely. But the main thrust of this new volume from the Center on Education Reform is to abolish results-based accountability for charter schools and scrap careful vetting of would-be charter operators. Instead, they would rely on a marketplace free-for-all in which pretty much anyone can start a school and authorizers don’t shut (or non-renew) a school just because nobody is learning anything in it. “Standardized testing” is damned over and over again in these pages as if it were the root of all evil in today’s charter sphere.

This is a version of the familiar libertarian stance on charters (and school choice more broadly): the market will provide all the quality control that’s necessary. Quality is in the eye of the beholder, i.e., the parent—and the school operator. The heck with school outcomes.

This is idiocy. It’s also entirely unrealistic in the ESSA era. It arises from the view—long since dismissed by every respectable economist—that education is a private good and the public has no interest in an educated citizenry. Once you conclude that education is also a public good—one whose results bear powerfully on our prosperity, our safety, our culture, our governance, and our civic life—you have to recognize that voters and taxpayers have a compelling interest in whether kids are learning what they should, at least in schools that call themselves “public.” (For now, let’s set aside the thorny question of private-school accountability in an era of vouchers, tax credit scholarships, and suchlike.)

Test scores are by no means the only legitimate measure of whether kids are learning, and no state is still using “standardized” tests in the old-fashioned “norm-referenced” sense. In fact, a lot of today’s assessments are pretty darn good, and well aligned with the academic standards in which states set forth the essential skills and knowledge that they believe the next generation of their citizens should master. Nor is any state or authorizer worthy of its salt just looking at proficiency scores. They’re also looking at various gauges of student growth, as well as graduation rates, pupil and teacher attendance and persistence, and more (e.g., Advanced Placement scores, dual credit results, where kids go to high school after leaving the charter middle school, etc.). Good authorizers do site visits and pay attention to school climate.

To discard indicators of student learning, however, is profoundly unfair to the students themselves, particularly to poor kids and those with families that depend on other responsible adults and governing bodies—not just obscure school report cards on state websites—to winnow the choices so as to ensure that the schools they’re choosing among are worth choosing among. It’s also a raw deal for taxpayers.

I sincerely wish that every parent was a sophisticated school chooser and that the charter sector would produce more good choices for them, while winnowing out the bad ones. If it doesn’t do this via its own mechanisms such as authorizing, more states will pass clumsy “sudden death” laws to get rid of persistently low-performing schools (and you can bet that those death sentences will be based mainly on proficiency scores).

Are these folks really prepared to just hand out charters after a cursory screening? And just trust unproven people with our taxpayer dollars and our kids—after all that we've seen in Ohio and elsewhere, despite all that we know about greedy and sometimes criminal behavior in the charter space, despite mounting evidence of for-profit operators opting for shareholders over schoolchildren? And are we really supposed to pretend that test scores don't matter? Of course let’s consider other elements when judging a school’s success—including whether anybody wants to attend it. But are we really supposed to ignore immediately-measurable outcomes? Can any sane person picture waiting until an elementary school’s graduates are twenty-five to determine whether it succeeded with its students?

None of this is to justify the regulatory overburden that today plagues most charter schools and authorizers. None of it justifies clumsy, bureaucratic, paper-chase practices by authorizers. Certainly, none of it justifies the tendency of many states to substitute “compliance” with dozens of input-and-process rules for sophisticated attention to whether a school is delivering the results it promised, including (nearly always) results that also fulfill state standards.

Yes, the charter sector needs a regulatory overhaul. But after showing that the bathwater is indeed dirty (and slipping in a bit of welcome clean water), Jeanne and her team of smart analysts are at grave risk this time of drowning the baby. 

Those not familiar with the history of the social esteem fad of the 1980’s should read Checker Finn’s brief and appropriately scathing review of it over at Education Week. Finn’s right on target in asserting that self-esteem, as a catalyst for improving children’s academic achievement and a remedy for social ills, such as crime and substance abuse, was at best oversold and at worst deliberately misrepresented, in terms du jour, as being “evidence-based.” Finn’s on more shaky ground, however, in attempting to draw a line between the self-esteem movement of the 1980’s and the current, increasingly prominent field of social-emotional learning (SEL).

The social esteem movement was centered around making children and young people feel better about themselves. It sounds nice, but it had insidious effects on attempts to boost academic achievement and build competent and high-functioning young adults. It, in effect, became an argument against delivering any news to students that they were anything but absolutely wonderful and perfect. It became better to find something nice to say about a student essay rather than point out spelling or grammatical errors, preferable to stress the effort a student made on a math problem rather than to point out that the methods used were not ones that would lead to finding the correct answer. You can still see offshoots of that mindset all over the education sector, wherein, for example, very few candidates in teacher preparation programs earn less than a B in their coursework and everyone struggles to find a way to convey to a teacher or to a school that there’s room for improvement in [you name it; fill in the blank] without that appraisal being perceived or portrayed as an indictment of complete and utter failure.

In contrast, social-emotional learning is primarily about teaching students how to make better decisions and manage negative emotions. Rather than an attempt at making students feel better, it actually requires that students be fully aware that “happy” is not a perpetual state of mind and that anger, frustration, and disappointment are part of life’s grand deal. The solution isn’t trying to ameliorate those bad feelings through superficial affirmations from one’s self or others but rather to not let them lead to impulsive behavior or irrational decisions. Those are ideas firmly grounded in cognitive psychology for which there is a solid base of research of effectiveness with regard to promoting high functioning, adaptive behavior, and preventing or treating disorders such as depression and substance abuse.

I admit to some bias here because in a past life, as a postdoc at Yale University involved in the New Haven Public Schools’ Social Development Program, I worked for and closely with Roger Weisberg and Tim Shriver, two of the cofounders of the Collaborative for Academic, Social, and Emotional Learning (CASEL), which is the target of some Finn’s sharpest criticisms. This also means I observed their work in action on more or less a daily basis. Long story short, I saw an incredibly well thought-out approach that was grounded in research and informed by ongoing data collection that indicated it was changing the lives of its intended beneficiaries with none of the naïve sentimentality or simplistic, one-dimensional conceptualizations of humans that were part and parcel of the self-esteem movement.

This isn’t to say that some of Finn’s concerns aren’t valid. There’s the same danger of overpromise in SEL that there was with the self-esteem fad, despite the fact that the former, unlike the latter, is built on a solid research foundation. Too many school systems will, unfortunately, implement it badly, and that needs to be guarded against, monitored, and remedied—a necessary process with regard to any new education policy undergoing rapid adaptation, as SEL seems to be now. I disagree with many of those in the SEL field that it should be a major indicator in state accountability systems, and wonder if the field is fully aware that some promoting SEL merely want to deploy it as the newest tool to paper over low academic achievement. None of that means that SEL is phony or, as Finn asserts, a “hoax.” Done well and in conjunction with other policies, it has the potential to help students manage their lives, negotiate complex social pressures and temptations, and achieve their long-term goals, academic and otherwise.

Charles Barone is the policy director at Education Reform Now & Democrats for Education Reform.

Editor’s note: This article was originally published by Education Reform Now.

The views expressed herein represent the opinions of the author and not necessarily the Thomas B. Fordham Institute.

On this week's podcast, special guest Noelle Ellerson Ng, an associate executive director at the AASA, joins Mike Petrilli and Alyssa Schwenk to discuss how to bridge the divide between education research and policy. During the Research Minute, David Griffith examines the recent CREDO study of charter management organizations.

Amber’s Research Minute

James L. Woodworth et al., “Charter Management Organizations 2017,” Center for Research on Education Outcomes (June 2017).

More than forty states got waivers under the Obama Administration, in part to get around NCLB’s unrealistic expectation that all schools would be proficient by 2014, but the states had to promise aggressive reform efforts in return. Studies that examine the impacts of some of the key provisions of this policy are starting to trickle in—and one such study of Kentucky was conducted by Stanford’s Tom Dee and colleagues. Recall that under the waivers, the feds required that states identify schools where subgroups of students have the lowest achievement. These were to be known as “Focus Schools,” and were to implement “research based interventions.”

The Bluegrass State is interesting because it was the first state to adopt the Common Core, and it won $17 million in the federal Race to the Top competition. It was also among the first group of states to apply for a federal waiver from NCLB. It developed explicit guidance for Focus Schools (more on that below) and used a “super subgroup” measure that combined all traditionally low-performing subgroups—meaning those who were eligible for free or reduced price lunch, in special education, black, Hispanic, American Indian, and English language learners—into an umbrella group, which they called a “student gap group” or SGG.

In Kentucky, all schools must have a “comprehensive school improvement plan” overseen by a school committee—comprising parents, students, and community members—in charge of designing and monitoring the plan. The Focus Schools, however, have additional responsibilities, including ensuring curricular alignment to the Common Core, providing time for teacher collaboration and data-driven strategies, and devising strategies to improve school safety and discipline.

In the study, analysts use school report card data for over nine hundred elementary and middle schools, school-level measures of student proficiency, and various school-level data on schools, teachers, and students. They focus the analysis on the first full year of implementation (2013–14) because the additional requirements, originally required only of Focus Schools, were extended to all schools after that year.

About one-fifth of the analytic sample (187 schools) had Focus School status. The student gap score, based on the 2011–12 state test results, is the main criterion that determines Focus School status. The cutoff for inclusion as a Focus School is specific to each school level; schools with an SGG score in the bottom 10 percent of elementary or middle schools are eligible. Analysts use a regression discontinuity design where they compare the “gap groups” in schools just above and below the cutoff.

The analysts find substantial improvements in school performance. Specifically, the Focus Schools saw math achievement rise by 5 percentage points, which is equivalent to a 17 percent increase for the targeted students compared to the comparison group mean. Reading achievement increased by 9 percent. They also find positive but not always statistically significant results for the non-gap group.

Other state studies of reforms under waivers (like in Louisiana) are not finding differences like these. Analysts hypothesize that designating a large, umbrella group of underserved students (a.k.a. a super subgroup) helped to foster improvement more so than narrowly targeting reform efforts on one smaller subgroup. What’s more, the study established that such a group was not masking low individual subgroup performance. That’s super news for super subgroups.

SOURCE: Sade Bonilla and Thomas Dee, “Sade Bonilla and Thomas Dee, "The Effects of School Reform Under NCLB Waivers: Evidence from Focus Schools in Kentucky," Center for Education Policy and Analysis (June 2017).

A recent survey in Ohio offers, from an educator’s point of view, insights on standards implementation that are applicable in the other forty-nine states and D.C. In spring 2016, researchers from the Center on Standards, Alignment, Instruction, and Learning (C-SAIL) surveyed 417 teachers along with 153 principals and administrators working in forty-two Ohio school districts. The survey explored three key questions: What are the most significant implementation challenges? What resources are needed to implement the standards? And are Ohio’s learning standards in math and English changing the focus of instruction?

Teachers cite time constraints as a significant implementation challenge. A majority of teachers (54 percent) say that insufficient class time is a moderate or major challenge, while 41 percent report a lack of planning time. Teachers view these time crunches as greater challenges than other organizational concerns, such as staff turnover, class sizes, or inadequate school resources. Meanwhile, principals view “inadequate lead time to prepare for implementation” as their biggest challenge. Both teachers and administrators note considerable challenges with the wide range of student abilities and the lack of parental involvement, though it’s less clear how exactly these relate to standards implementation.

When it comes to helpful materials, educators generally agree that curriculum resources aligned to standards are of significant value. Teachers also cite digital tools and professional development as important. Meanwhile, both teachers and administrators rank standards-aligned textbooks as the least useful implementation resource. What educators may be implying is that curriculum resources (such as Engage NY) are in and textbooks are out.

The survey finds that some, but not all, teachers are shifting their instructional focus to the content or skills emphasized in the new standards. Ohio’s elementary math teachers report strong coverage of the emphasized content. However, elementary ELA and high school math teachers say that they actually cover more deemphasized content. High school ELA teachers report roughly equal focus. The researchers cannot pinpoint why these patterns exist but it’s worth noting that they are not unique to Ohio; these patterns in content coverage mirror C-SAIL survey results from Kentucky and Texas. This finding has led C-SAIL analysts to conclude that “states and districts could provide more support in helping teachers move away from certain content.”

Raising Ohio’s academic standards has been a crucial school reform. While educators have had the better half of a decade to implement these more rigorous standards—they were first adopted in 2010—this survey describes some of the ongoing implementation challenges. Ohio policymakers—especially those who would have the state shift to yet another set of standards—should keep in mind that proper implementation is not as simple as flipping a switch. And those in other states would be wise to heed this lesson, too.

Sources: Laura M. Desimone, Adam K. Edgerton, and Rui Yang, “Standards implementation in Ohio: Local perspectives on policy, challenges, resources, and instruction,” C-SAIL (March 2017); and Adam Edgerton, Morgan Polikoff, and Laura Desimone, “How is policy affecting classroom instruction?” Brookings Institution/C-SAIL (May 2017).

The Archbridge Institute has kicked off a three-part series that explores intergenerational economic mobility—i.e., how much people’s income differs from that of their parents. In the first installment, author Scott Winship attempts to make sense of what he calls an “explosion of mobility research.”

The report uses data from the Panel Study of Income Dynamics (PSID), which has tracked the income of a nationally representative sample of adults and their children for nearly fifty years. Given income’s sensitivity to age and chance, assessing economic mobility is challenging. To address this, data are collected from over six hundred parents and their children starting at age forty (or as close as possible to it), the age that has been found to be most representative of lifetime income. Then, bi-yearly income averages for parents and their children are calculated between the ages of twenty-five and fifty-five.

Distributional analyses show that absolute intergenerational mobility is around 75 percent, meaning that about three out of four children will grow up to make more money (adjusted for cost of living) than their same-sex parent. Economic growth is encouraging but does not always translate to opportunity and access because it does not factor in “how well or poorly peers have done.” For instance, if everybody’s incomes increase by the same amount, there will be no change in people’s ranks; the bottom may be higher, but it is still the bottom. Assessing how a person’s economic ranking compares to that of his or her parents requires a different measure—relative intergenerational mobility.

Relative to their peers, the poorest and richest children will likely grow into the poorest and richest adults. Children of parents in the middle experience markedly higher levels of mobility in both directions at nearly balanced rates, meaning it’s equally likely for them to move up a rank as it is for them to move down. In sum, per the absolute measures, most people are making more money than their parents, but few are upwardly mobile. Overall, Winship’s findings suggest that obstacles to overcoming disadvantaged family origins remain large, and that “past research has overstated the extent to which patterns of relative mobility reduce childhood relative income gaps.”

The author highlights several important limitations to the study. First, he stresses the mobility estimates presented are descriptive, and not causal; it is very likely that many factors other than parental income affect children’s life trajectories. Second, PSID’s sample is fairly small and, due to data availability, income amounts used in the study are multiyear averages. Winship also stresses that, if given access to data that spans more years, mobility levels would be even lower because outliers would be less impactful.

So, what is education’s role in this discussion? The report focuses on economic over educational mobility measures, the author argues, because educational attainment is less reliable than income for several reasons. For one, the range of educational attainment is limited (and crowds at twelve and sixteen years of schooling). Additionally, the amount of educational attainment does not always speak to the quality of the education received. Therefore, it is harder to establish relationships between education and other life outcomes. Though compelling, other research does make this link.

Low-income students make up nearly half of the U.S. public school population. In line with Winship’s findings, it’s likely that many of those students will eventually raise their own children in the economic conditions they were raised within, and education is a key variable in this equation. This is significant for states that are completing their ESSA accountability plans. Intentionally targeting disadvantaged students, including those that are higher achieving, may help close the gap.

SOURCE: Scott Winship, “Economic Mobility in America: A State-of-the-Art Primer,” Archbridge Institute (March 2017).