Study after study has found that urban charter schools, and non-profit charter networks in particular, tend to be more successful at boosting student achievement than traditional public schools in similar settings. But why?
Study after study has found that urban charter schools, and non-profit charter networks in particular, tend to be more successful at boosting student achievement than traditional public schools in similar settings. But why?
In recent years, the Thomas B. Fordham Institute has published several studies that provide partial answers to this question. Student-Teacher Race Match in Charter and Traditional Public Schools, by American University’s Seth Gershenson, found that Black students in North Carolina charters are about 50 percent more likely to have a Black teacher than their traditional public school counterparts—which may help to explain why Tar Heel State charters are more effective at boosting the achievement of Black students. Similarly, Teacher Absenteeism in Charter and Traditional Public Schools, by Fordham’s David Griffith, found that teachers in traditional public schools were almost three times as likely to be chronically absent as those teachers in charter schools.
Yet one riddle continues to perplex: We know from copious research that new teachers tend to be less effective than educators with more experience. But despite having a higher proportion of junior staff, urban charter networks often outperform their district peers.
So are charter school teachers just higher performing to begin with? Or are charter schools and/or charter management organizations (CMOs) particularly effective at helping new teachers improve faster?
Although there are plenty of well-known differences between charter and traditional public schools, very little is known about how teachers in charters improve over time—a gap that seems well worth filling, since schools of all types should be able to learn from one another when it comes to teacher recruitment, development, and retention.
Toward that end, for our new report, Teacher Effectiveness and Improvement in Charter and Traditional Public Schools, we invited George Mason University associate professor Matthew Steinberg and University of Pennsylvania doctoral student Haisheng Yang to dig into how teacher effectiveness varies and evolves across traditional and charter public schools, as well as within the charter sector. Dr. Steinberg has published a number of studies on teacher quality and effectiveness, finance reform, and school discipline, including Fordham’s first report on discipline policy. He and Yang have also collaborated to examine principal mobility and professional development in Pennsylvania.
In Teacher Effectiveness and Improvement in Charter and Traditional Public Schools, they use data on teachers who worked in Pennsylvania’s charter and traditional public schools between 2007–08 and 2016–17 to investigate teacher effectiveness, improvement, and mobility across and between sectors. Pennsylvania is an ideal setting for such an investigation because the growth of its charter sector mirrors national trends, and because it has a healthy supply of both CMO and stand-alone charter schools. Accordingly, Steinberg and Yang examine academic outcomes for math and English language arts (ELA) for all students in grades three through eight, including those enrolled in both types of charter schools.
We strongly encouraged you to read the full report, which is thought-provoking, concise, and full of interesting statistics and figures. But for now, here’s a summary of the key findings.
According to Steinberg and Yang, these findings suggest that Pennsylvania’s CMOs are succeeding with “a fundamentally different approach to human capital” than the state’s traditional public schools. But how exactly are they managing that?
In our view, the study offers evidence for two potential drivers. First, the second and third findings suggest that CMOs are better at systematically identifying and promoting their most effective teachers to instructional leadership positions (e.g., literacy coach or master teacher) where they can help other teachers improve—and by extension, help more students.
Second, the data indicate that CMOs are more likely to part ways with their least effective teachers. In fact, over 30 percent of CMO teachers exit their schools annually—voluntarily or involuntarily. Yes, that’s a lot of turnover, and some of it is surely due to tough working conditions (including high expectations for teachers) and low pay (due to the substantial gaps in per-pupil funding between the sectors). Still, insofar as it’s attributable to the exit of ineffective teachers, higher turnover is a feature rather than a bug. And insofar as teachers who don’t improve are more likely to leave, higher turnover might also explain why teachers who stay in CMO-run schools improve so quickly
Of course, it’s also possible that CMOs are doing a better job of recruiting smart and highly motivated teachers who are likely to improve more quickly. For example, one “School Leader’s Toolbox” published by The New Teacher Project (now TNTP) urges schools to “define the Ideal Teacher...based on what type of teacher has been successful in your school” and to use the hiring process to set expectations by “communicat[ing] what is exciting and challenging about working at your school so that candidates are prepared for the school’s culture and unsuitable candidates self-select out of the process.” Perhaps CMOs do a better job of following this advice.
In addition to these explanations, there are other possibilities (though, without data, we can only speculate). For example, prior research finds that CMO principals are more likely than their district peers to (1) report that their schools define and enforce a comprehensive set of behavioral standards, and (2) require parents or students to sign an agreement acknowledging their responsibilities. So perhaps these common, schoolwide expectations around behavior help new teachers learn to manage their classrooms more quickly, allowing them to focus on their craft.
Or perhaps CMOs’ practice of deploying a common curriculum makes it easier for their new teachers to improve. As our colleague Robert Pondiscio has written, a schoolwide curriculum fundamentally changes the nature of a teacher’s job, especially if he or she is new to it. Rather than spending countless hours fumbling in the dark—or at least on the Internet—for instructional materials, she can spend that time studying lesson plans, building relationships with families, and offering feedback on student work. “Expecting teachers to be expert pedagogues and instructional designers,” explains Pondiscio, “is one of the ways in which we push the job far beyond the abilities of mere mortals.”
Each year, American schools hire approximately 200,000 new teachers. And because turnover rates have increased, 38 percent of K–12 teachers in the U.S. now have less than ten years of experience. These numbers surely make the case for doing more to retain great teachers, but they also show just how important it is that we help new teachers get better faster. As it turns out, accelerating teacher improvement is yet another area where all schools could take a page from the success of our nation’s most effective charter school networks.
“In the midst of winter, I found there was, within me, an invincible summer.” —Albert Camus
By now you’ve heard the ominous warning issued by President-elect Joe Biden that the country is facing a “dark winter.” For many school districts across the nation, this week is the last one before the holidays, and most large ones are only offering remote instruction. The lack of constitutional authority notwithstanding, Biden has promised to safely and responsibly reopen the “majority” of schools for in person learning within his first one hundred days in the White House. Practically speaking, this would entail swinging open the physical doors on at least half of the nation’s 130,000-odd schools by the beginning of May, when many schools ordinarily prepare to end the school year.
What this timeline portends, especially for the majority of large districts that have intermittently closed or remained physically shuttered, are more enervating stories ahead about learning loss and missing students, as well as harrowing accounts of child abuse, adolescent depression, and suicide.
The daunting outlook means things may well get worse before they get better, but it hasn’t been all gloom and doom. Indeed, there have been genuinely encouraging developments that provide some reason for Yuletide cheer in a year that’s been otherwise wanting for good news. Five trends in particular could pay dividends when the country finally kicks this godforsaken plague to the curb.
First, parents cannot unsee their children’s virtual lessons, no matter how earnestly some districts may wish. This newfound transparency has been an adjustment for teachers and parents alike, but the dynamic has empowered more parents and families to take control of their children’s learning. From homeschooling to learning pods, a burgeoning parent activism could exert some long overdue pressure on states and districts to diversify their delivery system and design programs that are more accommodating to the needs of students and families. One stellar example of this is a charter school in New Jersey that has started offering evening classes for Kindergartners to help families coping with Covid-19.
Second, the “whatever it takes” ethos behind efforts like the one just mentioned require courageous leadership. Almost axiomatically, state and district chiefs have in too many instances made reopening decisions that are convenient for the teacher unions and the bureaucracy, leaving students and families out in the cold. In contrast, consider this from a recent op-ed by former Dallas superintendent Mike Miles:
My network, Third Future Schools, has conducted in-person learning since Augugt 5 (after six weeks of in-person summer school), and have had relatively few teachers and students test positive and virtually no transmission within any of the schools. We will likely remain open through this pandemic because mitigating learning loss and growing student achievement will positively impact our students in both the short and long term, while the failure to provide students with first-rate instruction during this time will negatively impact them for years to come.
Education decision-makers would do well to adopt Mike’s clarity of vision and purpose as they regroup and recalibrate for the second half of the school year.
Third, charter schools have been on the ropes nationally, still ducking and weaving in states to avoid every blow. To wit, when Covid-19 upended routines for millions of students last spring, leading charter school networks sprang into action to significantly improve the remote learning experience for students, teachers, and families. Their response to the viral challenge offers many lessons. Among them, the importance of leaning into a team approach to curriculum and instruction, recreating the structure of the normal school day and rigorous grading practices, and reaching out to individual students and families on a regular basis.
Fourth, the crisis has highlighted the need for common, high-quality curricula because charter networks and districts with one in place were able to more easily pivot to remote or hybrid learning. To be sure, the critical conversation on the science of reading was briefly sidelined in favor of getting virtual instruction right, but it has since found its legs again. In August, Emily Hanford published her latest entry on America’s approach to reading instruction, and then two months later, got a hold of internal documents in which self-proclaimed literacy guru Lucy Calkins partially walked back her influential curriculum and philosophy for teaching English language arts. There’s plenty of reason for skepticism, but the ample energy and attention being given to parental demand for scientifically sound practice is momentum worth sustaining.
Finally, high-dosage tutoring is having a moment here and overseas. The evidence behind it is clear and compelling, though the problem of scaling is difficult to overstate. Still, assuming a sizable federal relief package, there’s been talk about incentivizing states to go gangbusters on summer school next year when—with logistics and luck—an effective vaccine will have been widely administered. The nation would then pull out all the stops in the fall to reap the benefits of small-scale instruction. If such a post-Covid Marshall Plan comes to fruition, we would hopefully pull millions of students back from the brink.
As bad as things have been, they likely could have been much worse. The creativity and resilience of gutsy educators and plucky parents throughout this crisis—all the while as national and state policies treated education as an afterthought—have surely kept more students from sinking. If Biden wants to “build back better,” these five promising trends can help provide a sturdy foundation. Taken together with the astonishing pace of progress on the vaccines, which are nothing short of miraculous, schools and systems could have something to look forward to after the page is turned on the year of the Lord 2020.
Editor’s note: This is the first post in a five-part series about how to effectively scale-up high-dosage tutoring. Read parts two, three, four, and five.
High-dosage tutoring (HDT) is “having a moment”—in the USA, Netherlands, and UK. See here, here, here, here, here (UK), here, and here at Fordham.
Andrew Rotherham, however, weighs in with a cautionary note. He writes:
If you invest in the silver bullet market there is a buy opportunity coming in tutoring. Not just any tutoring, high-dosage tutoring. The word itself sounds exciting—high-dosage!
Here is how these things tend to go. New idea—or not new but reintroduced idea—widely implemented through a funding and think piece gold rush. And widely implemented in uneven ways with little fidelity to the research because of the haste and good intentions coupled with lack of capacity around the field. End result, good idea gets discredited because, on average, it shows little if any impact. You see this around the ed tech sector, class size reduction, teacher evaluations, some reading initiatives, charter schools.
Andy nails it.
This five-blog series is about our shared fear that HDT will scale up badly.
Some quick background: We met on a basketball court thirty years ago. We’ve chased two education reforms in our careers: school culture and high dosage tutoring. We each penned books about school culture, with Mike outselling Bo eight copies to six copies (two more cousins). No traction there.
When it comes to High Dosage Tutoring, however, we’ve been part of some successes, in the USA and in the Netherlands.
Mike’s effort in Boston—Match Education—achieved large gains. Harvard’s Roland Fryer led a replication of HDT in Houston. Alan Safran of Saga Education improved and scaled the program in Chicago and NYC and beyond. Mike Duffy and Jared Tailefer turned the idea into Great Oaks, a HDT-fueled charter school network in cities like Wilmington, Newark, and Bridgeport.
The Amsterdam program, where Bo was an advisor, had students, after half a year, making math gains between 0.43 and 0.70 standard deviations—huge gains. The Bridge Learning Interventions’ version of HDT has been replicated in multiple Dutch cities, and is now scaling up.
We are convinced high dosage tutoring can work. Often does work. HDT has huge potential. But since every other edu-idea that’s been scaled has failed, is there any way to avoid that fate here?
Hell if we know! We’re terrible at politics.
Our contribution is an effort at an explainer on why HDT will not respond well to a “just put money behind it” policy effort.
John Arnold puts it this way:
And to explain the tutor scale-up problem, we turn to an apropos analogy: Vaccine development.
This next four blogs in this series are:
“High-dosage tutoring” has become the common parlance. But we mean “high-impact tutoring,” meaning that irrespective of dosage, kids actually made large learning gains when measured in a randomized control trial.
Final note: Mike already co-wrote a cautionary essay for the Brookings Institute on this very topic. But the final product wasn’t cranky enough for his taste. That’s because his co-writer, economist Matt Kraft, a dear friend, is much more optimistic about scaling HDT.
So, Mike turned to Bo, who shares Mike’s (and Andy’s) glass half empty spectacles. Together, we’ll explain why HDT is quite hard to scale, and describe a narrow path to do it right.
Despite a stampede of interest in students’ social-emotional development (SED), gathering data on—and measuring the success of—such initiatives remains a challenge. Researchers often rely on self-reported surveys in which students respond to questions about their experiences, attitudes, and beliefs related to school and themselves. But can these be used to accurately determine whether some schools are doing better at promoting SED than others? If so, can attending such a school boost a student’s chance of postsecondary success? A recent study from researchers at the University of Chicago and Northwestern University suggests that the answers are “yes” and “yes.”
Using survey, administrative, and test-score data for six cohorts (55,560 students) of first-time ninth-grade students who attended a Chicago public high school between 2011 and 2017, the analysts examine the extent to which some schools do more than others to improve students’ social-emotional development in two major domains—social well-being and work habits—and how such schools impact students’ short and long-term outcomes.
First, to identify which schools were better at promoting social-emotional development, the researchers conducted a value-added analysis that sought to identify causal effects of individual schools on students’ test scores, social well-being, and work habits. Social well-being and work habits were measured using the 5essentials Survey, a survey that the Chicago Public School system has been administering to sixth- through twelfth-grade students since 2010. (Click here to see sample survey questions.) To conduct the value-added analysis, test scores and survey responses of each student in a particular school were compared to similar students in other CPS schools at the end of ninth grade. Students were defined as “similar” based on a broad range of criteria that included gender, ethnicity, the socioeconomic status of their census block, and free-or-reduced-price-lunch status. A school’s value-added was then determined by how much it increased students’ test scores and social-emotional development relative to measured changes in other schools. Schools were determined as having a high value-add in SED if the score in each domain—social well-being, work habits, and test scores—was 1.0 standard deviation higher than the average school in the study, or equivalent to attending a school in the 85th percentile of performance in a particular domain.
The researchers then used these data to measure how schools with high value-add in SED affected students’ test scores, as well as other short and long-term outcomes. These short-term outcomes included the number of absences, number of disciplinary incidents that made the student eligible for suspension, whether a student had ever been arrested on or off campus, and a “freshman on track” measure (i.e., whether a student earned five full-year course credits and no more than one F per semester of work in a core course in their freshman year). Long-term outcomes included high school completion and college enrollment.
Of the paper’s many results, one noteworthy finding is that schools that excelled at improving students’ social well-being and rigorous work habits also raised students’ test scores by 6.0 and 5.7 percent of a standard deviation, respectively. This important finding suggests that schools that have a strong history of improving test scores aren’t necessarily focusing on academic achievement alone; rather, strengthening students’ social-emotional development may help facilitate academic success. “Both/and” rather than “either/or” wins the day.
Also of note is that better social-emotional development appears to improve rates of high school graduation and college enrollment. For example, students who attended a high school that excelled in all three domains measured by the study—social well-being, work habits, and test scores—were 1.9 percentage points more likely to graduate high school, and 3.6 percentage points more likely to enroll in a four-year institution within two years of graduation, compared to students attending the average school. Meanwhile, schools with strong test score value-added alone only improved students’ likelihood of graduating by 1.2 percentage points and four-year college enrollment by 2.3 percentage points.
Why might schools that have a history of building students’ social-emotional development see better long-term outcomes for their students? Authors posit that this is because schools with high value-added scores in all three domains improve students’ “freshman on track” status. These schools are also better at reducing school absences (-1.3 days), decreasing disciplinary instances (1 percentage point), and reducing school-based arrests (21 percent) than the average school in the system. While schools with a reputation for high test-scores only did reduce these negative school experiences, they did so at a much smaller rate.
Whether these findings will affect test-based accountability policies aimed at boosting students’ academic success and postsecondary outcomes remains to be seen. If tests are a weaker predictor of these outcomes—and if schools that prioritize social-emotional development see improvement in both test scores and key long-term outcomes—instruments such as Chicago Public Schools’ survey could play a role in the “next generation” of assessments and school accountability. We could see systems incorporate broader measures of school quality that take into account multiple definitions of student development, taking some of the pressure off of tests alone. This study suggests that that would be OK—even for organizations like ours that see the merit in test-based accountability.
SOURCE: C. Kirabo Jackson, et al., “Linking Social-Emotional Learning to Long-Term Success,” Education Next (Fall 2020).
Like traditional public schools, charter schools are publicly funded according to student enrollment. But compared to their district counterparts, charters have long received far less per-pupil funding.
A new study is the latest to look at charter and traditional public school (TPS) funding in major metropolitan areas across the U.S. It explores whether charter and traditional public schools receive equitable per-pupil funding—and if not, what factors are causing the funding disparity.
In this latest analysis, researchers from the University of Arkansas examine state records and audited school reports for the 2017–18 school year in eighteen metro areas: Atlanta, Boston, Camden, Chicago, Denver, Detroit, Houston, Indianapolis, Little Rock, Los Angeles, Memphis, New Orleans, New York City, Oakland, Phoenix, San Antonio, Tulsa, and Washington, D.C. Specifically, they compare per-pupil revenues for public charter schools to traditional public schools within geographic city boundaries.
The analysis finds that on average, across all 18 locations, charter schools receive an average of nearly $7,800 less per-pupil than TPS, or 33 percent less in total annual funding. The authors put it starkly, “students in public charter schools sacrificed around one-third of their educational resources by opting out of their traditional public schools.” And this large gap has grown significantly over time. A prior analysis conducted for Fordham in 2002–03 found that, on average, charter students received $3,266 less in funding than their TPS peers, meaning that the charter funding gap has more than doubled in just fifteen years.
Public charter school students in Camden, New Jersey, were the most underfunded in terms of dollars, receiving an average of $16,317 less in per-pupil funding than TPS students, a 46 percent funding inequity. Charter students in Little Rock, Arkansas, were the most underfunded on a percentage basis; they received a whopping 57 percent less funding than their TPS peers, a difference of $11,327 in funding. Our home base of Washington, D.C., didn’t fare much better, with a revenue disparity of over $11,000 or 31 percent.
Of all the locations analyzed, only one, Shelby County (Memphis), Tennessee, demonstrated an equitable revenue balance between charters and TPS, with students attending charter schools receiving 96 percent of the per-pupil funding that their peers attending traditional public schools are allotted.
Might student composition explain the funding gap? To explore this question, the researchers conducted additional analyses adjusting for observable differences in student populations, specifically looking at whether charters and TPS serve the same numbers of students from low-income families, English language learners (ELL), or students who have special needs. Of these three variables, they find that only controlling for enrollment rates for students with disabilities significantly explains variation in per-pupil revenue. But two-thirds of the funding gap remained after controlling for measures of student disadvantage.
Finally, researchers explored what funding sources are contributing to public school funding inequities, including federal, state, local, public, and nonpublic funding sources. While these vary significantly between locations, they found that overall, “a dearth of education funding from local sources was most responsible for the charter school funding gap.” Federal education revenues also worsened the gap, as overall, charters receive 37 percent less in federal dollars per-pupil than TPS.
After nearly thirty years on the education scene, and now serving over three million students across the country, it’s clear that charter schools are here to stay. And even with lower levels of funding, urban charter schools in particular achieve great outcomes for low-income students and students of color. One can only imagine what type of results they might obtain in a world where both public school sectors were equitably funded.
Source: DeAngelis, Corey A., Wolf, Patrick J., Maloney, Larry D., May, Jay F. (2020). Charter School Funding: Inequity Surges in the Cities. Fayetteville: University of Arkansas, Department of Education Reform.
On this week’s podcast, Noelle Ellerson Ng, associate executive director at AASA, the School Superintendents Association, and Emily Oster, Professor of Economics at Brown University, join Mike Petrilli and David Griffith to discuss whether the vaccines will allow schools to reopen nationwide this school year. On the Research Minute, Matthew Steinberg discusses the study he and Haisheng Yang just conducted for Fordham on teacher effectiveness and improvement in charter and traditional public schools.
Matthew P. Steinberg and Haisheng Yang. Teacher Effectiveness and Improvement in Charter and Traditional Public Schools. Washington D.C.: Thomas B. Fordham Institute (December 2020). https://fordhaminstitute.org/national/research/teacher-effectiveness-and-improvement-charter-and-traditional-public-schools
If you listen on Apple Podcasts, please leave us a rating and review - we'd love to hear what you think! The Education Gadfly Show is available on all major podcast platforms.
