The Education Gadfly Weekly

It’s an experience common to every writer in the world: the sinking feeling when you realize that the provocative headline your editor has placed atop your carefully-wrought piece is sure to send an angry mob in your (virtual) direction. So it was with my recent Washington Post op-ed arguing that “millions of elementary pupils who were already a year or two behind when the crisis struck” should not be socially promoted next year, so as to give their schools and teachers the gift of time “to fill in the gaps in kids’ knowledge and skills and accelerate their progress to grade level.” The truncated, click-bait version? “Schools should consider keeping kids in the same grade this fall.”

As Jay Mathews wryly noted in a follow-up column, my commentary “apparently enraged nearly everyone who read it.... Reader comments included: ‘I’m rolling my eyes so hard it hurts’ and ‘My question is how this idiocy found its way into The Post.’”

Jay came to my rescue, saying (correctly) that my article was really “a plea for creative thinking”—especially regarding how schools might use time differently in the face of this unprecedented challenge. Bingo—and something I should have said more clearly myself!

The flap reminded me yet again that words and language matter, especially when we’re dealing with what David Tyack has called “the grammar of schooling.” Washington Post readers (and especially folks on Twitter who never got past the headline) saw me proposing to hold kids back, which struck them as mean, punitive, and “blaming the victim.” That of course was not my intention. I wanted to give struggling students something extra, something that would benefit them: not just an extended school year or day or small group tutoring (though we should do all that), but an extra, free year of schooling, if that’s what they need.

Lesson learned: We need to be precise, especially in this moment of heightened anxiety and widespread uncertainty.

So it is with the larger conversation about how to help “catch kids up” when schools return to in-person instruction, one hopes in the fall—a conversation that has also been riddled with imprecision. And it creates the impression of disagreement when really we are just talking past each other.

To overgeneralize, let’s view this complicated matter of what to do when schools reopen as two main camps that come at it from seemingly opposite perspectives. Let’s call one Team Personalized Learning and the other Team Grade Level Standards. 

In broad strokes, TPL wants to make sure that every student masters key skills before moving on. And if teachers find pupils in their classes who missed critical concepts years before, they want the flexibility to go back and spend a good deal of time filling in the gaps before trying to tackle grade-level material. Of course, much of Team TPL would love to blow up the whole grade-level thing entirely and let kids proceed at their own pace rather than based on their age.

In contrast, TGL members worry that too much “going back and filling in gaps” means tons of dreaded “remediation,” shown by research to slow down student progress rather than accelerate it. Instead, they want teachers to spend most of their time on grade-level work, providing supports to students to help them access that “grade level material.” And they sure don’t want to make kids repeat a grade, given the multiple research studies showing that it doesn’t do much good, and in fact can do much harm.

It might seem that TPL and TGL are at loggerheads with irreconcilable views.

But that’s only because we are still talking in generalities; we haven’t yet gotten precise.

Let’s fix that. To answer the question of “how to catch kids up,” we need to ask at least three questions. First, how old are the students we’re thinking about? Second, which subject? And third, how far behind are they? Because the best path forward is going to look very different depending on the answers.

Consider some recent commentary on this topic. The other day in The 74, two captains of Team Grade Level Standards—David Steiner and Dan Weisberg—made a great argument for their view using high school English as their example.

You are a tenth grade English language arts teacher in Urban City, USA. Your curriculum says teach George Orwell’s novel 1984, but half your class lacks the vocabulary and interpretive skills to read the book. So, following your training to differentiate your teaching, you ask those students to read Lois Lowry’s The Giver instead—a seventh-grade text you think they’ll be able to manage…

Imagine that instead of assigning half the class The Giver, the teacher decided that no student was going to miss out on reading 1984. That teacher would have to do focused work with the less-prepared students: They might be asked to watch the movie rendition or listen to the book on tape; they might need to review plot guides or digital editions of early chapters with embedded vocabulary help and synopses. The key is that these interventions happen just before the whole class encounters the text. Done right, they can give students who’ve fallen behind access to grade-level materials at the same time as their peers.

Persuasive, yes?

Meanwhile, one of the captains of Team Personalized Learning—Joel Rose—has been making a strong argument using middle school math as his example:

Imagine that you’re a sixth-grade math teacher. It’s the first day of school, and the vast majority of your students arrived multiple years behind where they should be. Your job is to teach them concepts such as understanding percentages and dividing fractions. Both will appear on the sixth-grade state test, but your students never successfully learned much of anything about the basics of fractions in fourth and fifth grade.

What would you do?

Would you try to go back and address their unfinished learning from prior years and run the risk of not having enough time to cover all of the sixth-grade-level material? Or would you march dutifully through grade-level content and try to fill in gaps where you can, knowing that it probably won’t be enough?

Compelling, no?

Isn’t it possible that they’re both right? (Indeed, Weisberg and Rose co-authored a piece last year finding much common ground, and arguing that a combination of grade-level work and “filling in the gaps” is often the right move.)

Isn’t it likely that it depends on the subject matter and which specific knowledge and skills we’re addressing?

Given that there is a clear progression of standards in math, it makes a ton of sense that students will struggle if their understanding of prerequisites is faulty—especially when gaps are enormous. No amount of “supports” and “scaffolding” is going to magically make that problem go away. So we should encourage teachers to go back and help kids fill in the holes—while also helping students make progress on grade level material.

But English language arts—and especially reading—is different. After the decoding stage, it’s not a “skill” at all, as my colleague Robert Pondiscio often explains. A high-school student who can’t read “at grade level” can almost surely decode the words. What he can’t do is comprehend novels and informational texts because so much vocabulary and background knowledge is unfamiliar to him. So building up his vocabulary and content knowledge is the name of the game—which talented teachers can do with the strategies that Steiner and Weisberg suggest.

And all of this looks very different for our youngest children. A kindergarten student who was on the verge of sounding out her letters in March is most definitely going to need help picking up where she left off come fall. Likewise when it comes to counting to one hundred. Not to mention all of the social and emotional skills that kids learn (or are supposed to learn) in kindergarten. 

So how should we help catch kids up? Let’s get precise.

As with all students, children in the earliest grades will need to be diagnosed with high-quality assessments, both the standardized variety and those developed by teachers who know them best. For those found to be way behind, we should give them—and their teachers—the gift of time. Rather than call it “retention,” let’s dust off my idea of a “second 2nd grade.” Kids currently in second grade will graduate to grade 2.5 (ideally with their current teachers); schools can spread out all of the skills in the K–2 standards across four years instead of three. If fancy universities promised their students a free fifth year of the college experience given the current circumstances, people would cheer. We should do the same by offering struggling kids a free extra year of elementary school.

For older students, it depends on the subject and the specific skills that need strengthening. In math, we should err on the side of mastery, and if that means spending some time on below-grade-level content, so be it. But in English language arts, as well as social studies and science, supporting students to access grade-level content should be the default. And of course, we should follow the lead of great schools and draw tools from both tool boxes via a mix of well-designed small group instruction, one-on-one tutoring, online acceleration and enrichment, and whole class discussions.

None of this is going to be easy, but we also need an approach that’s not impossible. We shouldn’t pretend that the typical school is suddenly going to be able to embrace instructional models that are brand new to them, given everything else they are dealing with. But we also shouldn’t waive our hands and make-believe that kids will be fine if we just do what we always do. Specific, proven, precise strategies are the key to finding our way through this historical challenge and doing right by our children.

This year’s holiday from federally-mandated end-of-year assessments in math and English language arts will undoubtedly embolden test haters to declare once again—and louder than ever—that we never needed those damned exams in the first place and that our schools and students are far better off without them. Thus will commence another skirmish in a war on testing that rivals the Seven Years War. 

As Lynn Olson and Craig Jerald point out in a thorough and perceptive report from FutureEd, statewide assessments were getting plenty of pushback pre-coronavirus. The pandemic is merely deepening their peril and intensifying the pushback. Their seventeen-page report is well worth reading, as is the short version that appeared on Education Next. Their chronology is spot on. They’re right about the intensity of the “testing backlash.” I assume they’re right that this year’s testing moratorium will cause more people to urge scrapping the whole irksome, onerous process. And I agree that tests would be a lot more popular with educators if they were demonstrably helpful in improving instruction, which waiting to year’s end doesn’t do much to accomplish.

Yet there may also be a fundamental shortcoming in their analysis, a great big issue that they allude to but don’t develop: What if the hostility toward testing is not, at bottom, about the tests, but rather about what the present testing regimen is mainly designed to do, which is to hold schools and educators (and sometimes kids) to account for their results? What if tests are the unwelcome messengers, but what’s really at stake is the message? I submit that if testing vanished but some other form of results-based accountability remained, educators would complain just as much—and work just as hard to recruit allies among parents and others to discredit them.

Think about it. Nobody likes to be held to account for their results, particularly when embarrassment, inconvenience, and unwanted interventions, possibly even the loss of one’s diploma or one’s job, hangs in the balance. Doctors and hospitals don’t much like it when their infection, mortality, or readmission rates are publicized and sometime lead to sanctions. Restaurateurs understandably hate it when the health department padlocks their bistros or a reviewer offers no stars. Competitive divers bridle when judges give 1.5 points to their triple backflips. It’s human nature—and the higher the stakes, the stronger the feelings.

It’s different, of course, when high-stakes accountability leads to bonuses, gold medals, five-star ratings, and 9.5 point dives. Everyone loves accolades. That’s human nature, too.

But the reason America got into results-based accountability in K–12 education wasn’t because more accolades were needed. It was because many schools were producing wholly unsatisfactory results, sometimes for everyone in attendance, sometimes just for subgroups within the school. We were, it was rightly said, a “nation at risk” because of those weak results.

That led to national goals, to statewide academic standards, to mandatory assessments in core subjects, and to complex regimens by which to evaluate the scores on those tests and the remedies to be applied when scores were low.

It did not, in theory, have to be tests by which schools’ results were gauged. But for a host of reasons—cost, convenience, security, the appearance of uniformity, objectivity, and a sort of fairness, etc.—standardized tests are what we ended up primarily relying on.

Of course we got carried away, particularly in clumsy efforts to use kids’ test scores to evaluate teacher performance. Of course we neglected other important elements of learning besides what’s readily tested and important elements of good schools besides academic achievement. Of course we muddled—and still do—the balance between achievement and growth. Of course we didn’t pay enough attention to the seemingly Sisyphean quest for tests that would be both “formative” and “summative.”

All true, all culpable. Perhaps still all fixable. But I submit that the “peril” in which state assessments find themselves, according to Olson and Jerald and others who have analyzed the pushback against testing, is not fundamentally about testing burden or the distortions it causes in curriculum, pedagogy, calendars, etc. It’s about results-based accountability for a system that’s producing unsatisfactory results. Educators trying to escape the accountability have resorted to a war on tests themselves and convinced many, many others—especially parents—that tests are the problem.

Think about it this way: Imagine that we start rating elementary schools, A–F or five stars to one, based on how well their graduates do in middle school. We judge them by middle school grades, discipline, etc., not by tests. And we find a way to adjust for “value added” so as not to penalize schools just because their pupils are disadvantaged.

Extend that thought experiment to high schools. Imagine that we find ways to gauge—and report—their effectiveness based on how their graduates do in college and the labor market, as glimpsed in a recent Mathematica study of Louisiana high schools. Then we sanction or intervene in various ways in the high schools whose graduates fare poorly.

Would F schools be any happier once their grade isn't derived by tests? Would unions complain less vociferously if states moved to intervene with tough love in one-star schools? How much are you prepared to wager?

There are ample good reasons to find additional real-world measures of school quality and effectiveness, but let’s not deceive ourselves that doing so will end the war on accountability—or make our schools any better.

I repeat: Tests are the messenger, and it’s the glum message they continue to convey about many schools that’s the problem. Shooting them down won’t cause a single child to learn more, a single inept teacher to do a better job in the classroom, or a single crummy school to improve.

Otherwise, Lynn and Craig got it about right.

Editor’s note: A version of this essay was first published by Education Next.

As thoughts start turning to reopening schools, there’s been no shortage of advice on what educators need to do to prepare and how they should go about doing it. One emerging piece of consensus is that schools may need to start the school year remotely as part of rolling closures triggered by new outbreaks. On its surface, altering scheduling, staffing, and other logistics to effect social distancing seem like straightforward operational functions, but like all options being thrown on the table, each possibility presents a cascading series of obstacles, many of which could prove insurmountable.

Thinking back to the elementary school I once led, Elm City College Prep, we never had to tackle the novel challenges facing schools today, but there were many things we could have adjusted based on what the experts are forecasting. The stepped-up cleaning, physical distancing, and additional accommodations (e.g., allowing vulnerable teachers to work from home) would have called for considerably more planning than ordinarily required, but these are largely solvable problems assuming an adequate provision of resources and the suspension or removal of undue regulatory and contractual red tape. Granted, these are no small assumptions, but the logistics of remote learning and social distancing are not the most difficult part of the back-to-school equation.

No, the trickiest part in all of this—one that’s been lost in the Manhattan Project–style effort to build a new delivery system—is the primacy of school culture. The coronavirus has disrupted the rituals and routines of schooling, but it could have been much worse. Indeed, in hindsight, schools were fortunate that in-person schooling ended when it did in March, as teachers already had two-thirds of the year under their belts to establish relationships and make meaningful connections. I’ve seen the benefits of this firsthand with my five-year-old daughter as she wraps the year up with her teacher on Zoom. But how do you do that well with a new set of kids and a new teacher if their initial interactions are all or mostly online? It would seem almost impossible to start a school year with elementary age students in front of teachers they’ve never met and don’t feel any relationship with or accountability to.

One potential solution was raised by my colleague Mike Petrilli: Have kids stay in the same grade with the same teacher instead of moving onto the next. The idea was widely panned by educators and the public writ large, though policymakers were more sympathetic. Folks may have reasoned that the stigma associated with redshirting would have been yet another burden placed on the shoulders of students who, through no fault of their own, were already behind grade level. But regardless of one’s feelings about social promotion, there is considerable merit to Petrilli’s argument about students returning to the familiarity of their current teacher, especially as a hedge against the struggles many low-income, low-performing elementary students will assuredly face this fall.

A recent study suggests the answer could lie with what is often referred to as “looping.” Unlike retention, which requires students to repeat the same grade, looping calls for the teacher to follow her students to the next one. As one elementary teacher describes it:

It was such a positive experience. One of the big advantages is starting in September hitting the ground running—you already know the kids and the things they did the previous year and the things they need to work on.

There are no silver bullets, but looping would seem to address several of the uniquely sticky problems created by the pandemic. First, it directly confronts the issue of student-teacher familiarity, a foundation for learning and all of the other salutary effects of schooling. Second, it’s a low-cost policy, which is an important consideration with schools and districts facing severe budget shortfalls. Third, research shows not only improved test scores, albeit modest, for kids who loop, but the benefits accrue more to students of color, the population most likely harmed by school closures.

To be sure, looping all elementary grade students could present challenges. For one, moving teachers between grades would require them to quickly learn a new set of content standards and familiarize themselves with new materials. Second, the amount of research on looping is relatively thin, so it’s unclear whether the benefits would persist if the practice were suddenly expanded. Third, looping will not be effective, and can even be detrimental, if students are assigned to an ineffective teacher for two to three consecutive years. There’s also a question of what happens to today’s fifth grade teachers if their students move to middle school

Yet the potential upsides clearly outweigh the negatives. Teachers and students are already stressed by the “new normal,” and things will likely get harder before they get easier. As schools turn to what will be required of teachers in the fall, they must avoid complexity at all costs and do everything they can to make the job doable within the capacity of existing staff. Rather than having teachers fight through a screen to connect with their students, let’s have them build naturally upon the face-to-face relationships they’ve already established. Besides, ensuring forward progress on academics will be daunting enough.

Although there’s wide variation in teacher effectiveness, research shows that educators can learn from their colleagues and in supportive professional environments. Nearly all such research looks at year-to-year improvements. But could teachers also improve over the short term by doing what many of them, particularly at the middle and high school levels, do every day—that is, teaching the same material multiple times to different classes of students? Or could the tedium of such concentrated repetition lead to instructor burnout and declines in both teacher and student performance? A new working paper from the journal Education Finance and Policy digs into these questions.

The study was conducted in the Netherlands at a university business school, so there’s no claim of generalizability to an American K–12 context. But the teaching repetition model under study is not that far removed from the typical workload of American middle and high school teachers. And as this new study is only the second randomized controlled trial of the effects of teaching repetition ever conducted, its findings are worth a look.

Analysts observe 731 different instructors who vary in seniority from postdocs to full professors and each instructor teaches, on average, 2.5 sections of each course over six school years from 2009–10 to 2014–15. Thus, the sample includes around 7,300 sections. Of those, 42 percent are an instructor’s first section for a given course in the day. Thirty-three percent, 19 percent, and 8 percent are an instructor’s second, third, and fourth sections, respectively. This is the variation they exploit to estimate the impact of teaching repetition. Specifically, they are interested in four outcomes: a student’s final grade, whether they dropped out of the course, students’ evaluations of the instructor, and students’ self-reported hours per week they studied for the course.

Students are randomly assigned to sections within a course conditional on scheduling conflicts, which arise for about 5 percent of student-course registrations. Because the type of instructor and the course subject are likely to impact outcomes, the study compares student outcomes within instructor-course combinations, with the main analysis relying on comparisons between students in an instructor’s later sections of a course to their peers in the first section that have the same course plan. The effect of this is to control for the times that students take the course, so that time-of-day effects are not mistaken for repetition effects.

The headline is that the researchers find little to no impact for any of the outcomes. In terms of student grades, the estimates for an instructor’s first section relative to their third and fourth sections is statistically insignificant. There are also small and insignificant effects on teaching repetition for the probability of students dropping the course. Likewise, there’s little evidence that teaching repetition leads to better teaching evaluations, nor are there significant differences in self-reported study time related to teaching repetition. When they look at impacts by prior teaching experience, researchers also find just tiny differences, such that inexperienced instructors are not benefitting from teaching repetition compared to their more experienced colleagues. Finally, they examine whether having a same-day break in between teaching multiple sections changes the pattern of effects they observed in the study—akin to a planning period for American high school teachers—and might make for improvements, but again, no meaningful differences.

The researchers conclude that teaching multiple sections of a course neither helps nor hurts teaching effectiveness. Instructors don’t appear to use their first section as a trial run for later sections and students in earlier sections are not at a disadvantage compared to peers in later sections. Moreover, it appears that if they are going to get better as a result of repetition, they are going to need more than a short break in the same day. They also point out that the courses under study were mainly those where the instructor served as a facilitator—with students working in groups and teachers circulating to assist and provide answers to student questions—so perhaps different patterns would emerge in classes with different formats.

On the one hand, this study should be somewhat encouraging because it indicates that repetition does not lead to tedium-induced performance lags. On the other hand, neither does repetition boost short-term performance. Maybe part of the problem is that teachers aren’t accustomed to making adjustments on the fly in response to what they’re observing in real time. Such a process is nurtured through a building-wide culture of dynamism and adaptability that is deliberate, labor intensive, and yes, likely draining. It is far from the norm in American K–12 schools. But studies like this one remind us that it doesn’t have to be that way. Repetition need not be rote; with more effort, it can be riveting.

SOURCE: Harold E. Cuffe, Jan Feld, and Trevor O’Grady, “Returns to Teaching Repetition – The Effect of Short-term Teaching Experience on Student Outcomes,” Education Finance and Policy Journal (February 27, 2020).

On this week’s podcast, John Bailey, visiting fellow at AEI, joins Mike Petrilli and David Griffith to discuss AEI’s new Blueprint for Back to School. On the Research Minute, Amber Northern and special guest Paul von Hippel, associate professor of public affairs at the University of Texas at Austin, talk about Paul’s recent work on year-round school calendars and their many effects, including those on summer learning and achievement.

Amber's Research Minute

Paul T. von Hippel, "Year-Round School Calendars: Effects on Summer Learning, Achievement, Parents, Teachers, and Property Values," retrieved from Annenberg Institute at Brown University (February 2020).

The Education Gadfly Show is available on Google Podcasts, Libsyn, Spotify, iHeartRadio, Apple Podcasts, TuneIn, The Podcast App, PodBean, Podbay, Chartable, Castbox, and myTuner Radio.