Foreword
By Aaron Churchill and Chad L. Aldis
Public charter schools are one of the most enduring—and successful—educational innovations over the past three decades. Freed from suffocating union contracts and distracting school board politics, charter schools can create distinctive programs, assemble teams of mission-aligned educators, and maintain a clear focus on student needs. Today, approximately 4 million U.S. students attend the nation’s 8,000-plus charter schools; in Ohio, roughly 120,000 students attend the state’s 340 charters.
In exchange for these freedoms, charters are expected to deliver stronger academic outcomes. When they fall short, accountability mechanisms come into play, measures rarely found in the traditional district sector. These include corrective actions required by state officials or their sponsor (a.k.a., “authorizer,” the entity that allows a charter school to operate), immediate funding reductions when dissatisfied parents disenroll their children, and permanent closure by their sponsor or directly by state action if they chronically underperform.
The autonomy-in-return-for-accountability model has effectively served millions of American youngsters. According to the most recent national study by Stanford University’s Center for Research on Education Outcomes, the average charter student makes greater academic progress each year than a similar student attending a district school.[1] In Boston, researchers estimate that its finest charters close the yawning black-white achievement gap by one-fifth to one-third each year. New York City’s famed Success Academy charter schools consistently register near-perfect proficiency rates on state exams.[2] Ohio also has success stories, including well-regarded charters such as Dayton Early College Academy, KIPP Columbus, and the United Schools (all proudly sponsored by our sister organization, the Thomas B. Fordham Foundation).
Other charter schools, however, have struggled and many have closed due to accountability pressures and often-related fiscal challenges. In the early days of Ohio’s charter movement, too many schools were hastily launched, and others suffered from weak accountability and severe underfunding. Urged by choice-friendly and quality-centered advocates, including the Fordham Institute, Ohio has undertaken a decade-long turnaround effort via two main policy levers—both aimed at driving higher student achievement, an outcome the present report evaluates.[3]
- Stronger state oversight of charter sponsors. Ohio has long permitted various entities—e.g., local districts, educational service centers, and qualified nonprofits—to sponsor charter schools. A multi-sponsor system promotes competition and innovation, but without rigorous state oversight, it also allowed the growth of weak sponsors and lax accountability in Ohio’s charter sector. Enacted in 2015, House Bill 2 strengthened the state’s fledgling effort to hold sponsors more accountable for the performance of their schools. The legislation established consequences—including revocation—when sponsors turn a blind eye to quality. This tougher oversight had an immediate impact, with the number of sponsors falling from 66 in 2013 to just 19 by 2022 (18 are active today). The system also encouraged sponsors to shutter low-performing schools. Between 2016 and 2022, just over 100 Ohio charter schools closed (out of nearly 400 at baseline), largely done through sponsor action.
- More equitable funding for high-performing charter schools. Charter schools in Ohio, as in most states, have long operated with far less funding per student than their district counterparts, making it far harder to maintain high-performing schools. With the accountability framework solidified, state leaders tackled this major challenge confronting the charter sector. First enacted through the biennial budget for fiscal 2020 and 2021, the Quality Community School Support Fund (QCSSF)[4] has provided much-needed supplemental aid to qualifying schools, ranging from approximately $1,000 to $3,000 per pupil, depending on year (see below for program details). The extra dollars have been a game-changer, narrowing the per-pupil funding gap between these high performers and their local district from an approximately 30 percent shortfall to 10 percent.[5] In 2025–26, 116 of Ohio’s 348 charters qualified for the program.
As noted above, the goal of these reforms is to deliver better learning outcomes for students. Conducted by Dr. Stéphane Lavertu, an Ohio State University professor and Fordham Institute research fellow, this study offers hard evidence that they have indeed moved the achievement needle. Two findings stand out.
First, the overall performance of Ohio charter schools—the brick-and-mortar variety—has markedly improved. In 2007, the average charter student was losing ground in math and reading compared to a similar student attending a district school. The reverse is true today. In 2025, the average brick-and-mortar charter student gained the equivalent of 25 to 35 additional days of learning in both math and reading versus her district-attending counterpart. While the quality of individual charter schools still varies, the average Ohio charter today provides a superior education compared to students’ typical district alternative.
Second, providing more equitable funding has bolstered achievement in the state’s high-quality charters. Consistent with an earlier report examining the program’s first four years,[6] this study continues to uncover positive impacts of the additional funds into fiscal 2024 and 2025. To better isolate the impact of the program funds, this analysis compares narrowly qualifying to nonqualifying charters, rather than charter to district students. When doing this, we find that the extra dollars received by charter schools are associated with student gains worth about 50 to 60 days of additional learning.
Thanks to a decade of reform led by policymakers keen to ensure that charter schools serve students well, Ohio’s charter pupils are reaping the benefits of a more responsible, more accountable, better-funded, and higher-performing charter sector. Although charter students make more progress than they would attending district schools, most charter students (largely from low-income families) still score below the statewide average: For instance, 45 percent of third-grade charter students achieved reading proficiency in 2024–25 versus 61 percent statewide. Given this still wide achievement gap, the work of helping every charter student reach their full potential remains unfinished.
What to do? On the policy side, state lawmakers must continue to insist on strong accountability measures even in the face of demands—sometimes, unfortunately, from the charter sector itself—to water them down. Lawmakers still have their work cut out to ensure that all charter students receive the benefits of fair and equitable funding—not just those fortunate enough to attend schools qualifying for the extra dollars. As we at the Fordham Institute have long contended, charter school students shouldn’t get shortchanged simply because of their choice in public school. Lastly, state leaders should continue to protect the autonomy of charter schools, while balancing such independence with expectations to implement high-quality curricula, achieve student attendance targets, and hire top-notch teachers knowledgeable in their content area.
State lawmakers, including the next governor, should build on Ohio’s legacy of school choice and local autonomy as well as its more recent moves toward charter accountability and equitable funding. A renewed push for excellent charter schools will further empower families and put a next generation of charter students on pathways to success.
Executive Summary
Charter schools (labeled “community schools” in Ohio law) are publicly funded but independently operated schools that serve students from kindergarten through grade 12. Since their authorization in 1997, Ohio policymakers have promoted charters as a way to expand educational opportunities for families—particularly those residing in underperforming school districts—while creating competition that incentivizes all schools to better meet students’ needs. But charters are not necessarily good schools, and lawmakers have enacted a series of reforms aimed at improving their quality by holding them accountable for results and increasing their funding.
This report examines the trajectory of Ohio’s charter school sector between 2007 and 2025 and considers the role that accountability and funding policies have played in shaping that trajectory. Using student-level administrative data and school-level spending information, the analysis compares the academic outcomes of charter students with those of similar students attending nearby district-operated schools. It also examines how recent funding initiatives—particularly the Quality Community School Support Fund (QCSSF)—have affected charter operations and student outcomes. To enable apples-to-apples comparisons, this analysis focuses on brick-and-mortar charter schools that serve a general student population, setting aside online “virtual” schools and those that focus on dropout prevention and recovery.[7]
The analysis indicates that Ohio’s charter sector has improved substantially. The average annual achievement gains of charter students initially lagged the gains of similar students in nearby district schools, but the average charter now produces substantially larger achievement gains. Recent funding reforms also appear to have strengthened the performance of the charter sector’s most effective schools. Here are the key findings:
Ohio charter schools now produce substantially larger academic gains than nearby district schools.
During the 2006–07 school year, the average charter student had annual achievement gains that were 0.05 standard deviations lower than those of similar district students. By the 2024–25 school year, the annual gains of the average charter student exceeded those of similar district students by approximately 0.06 standard deviations. This advantage equates to roughly 25–35 additional days of learning per school year—and the gains compound for students who attend charter schools for multiple years. Quality varies, however. During the 2024–25 school year, 26 percent of charters posted average achievement gains in English that were statistically significantly lower than those of nearby district-run schools, and 40 percent of charters posted lower average gains in math.
Accountability reforms likely contributed to the sector’s improvement.
Ohio’s automatic charter school closure law—enacted in 2006—requires persistently low-performing charter schools to shut down, removing schools with weak academic results from the sector. Additional reforms enacted in 2015 strengthened oversight of charter school sponsors and operators. Together, these policies toughened the consequences for poor performance and appear to have contributed significantly to the steady improvement in charter school outcomes over time. For example, during the 2023–24 school year, 96 of the 226 charter schools for which gains could be calculated demonstrated statistically significantly lower gains than nearby district schools in at least one subject—and 10 of these schools were no longer in operation during the 2024–25 school year. On the other hand, 130 of these 226 charters posted higher annual achievement gains than nearby district schools in at least one subject, and all were still in operation the following year.
Charter schools continue to operate with fewer resources than nearby district schools.
Charter schools generally spend far less per pupil than district schools serving students from the same communities, as they receive less public funding. Recent policy changes have narrowed this gap by directing additional funding to higher-performing charter schools via the Quality Community School Support Fund (QCSSF). Nevertheless, the average charter school still spends approximately $4,000 less per pupil than the district schools where they are located, and the average school receiving QCSSF funding spends approximately $2,500 less per pupil.
Charter schools bringing in additional revenue from the Quality Community School Support Fund produce exceptionally strong academic outcomes.
After one or two years in a QCSSF-funded school, students experience large test-score gains of approximately 0.3 standard deviations in math and 0.2 standard deviations in English, roughly equivalent to their getting an extra 100–130 days of instruction relative to their charter-school peers who did not attend a QCSSF school. These schools also employ teachers with higher salaries and lower turnover and absenteeism, and their students exhibit lower rates of chronic absenteeism. The above results do not speak to the effectiveness of providing QCSSF charters with more funding, however, since Ohio selects QCSSF schools because they are highly performing.
Additional funding accounts for a substantial share of these improvements.
To isolate the impact of additional funding from the impact of selection based on quality, the analysis compares charters that narrowly qualified for QCSSF funding with those that narrowly failed to do so. In other words, it compares schools of similar quality that did and did not receive the additional QCSSF funding. The results suggest that roughly half of the achievement advantage associated with QCSSF schools is attributable to the additional funding, whereas the reductions in chronic absenteeism are primarily driven by the additional resources. Finally, the results indicate that QCSSF funding led to higher teacher salaries, lower teacher turnover, and lower teacher absence rates, providing clues on how funding might translate to improved outcomes.
The average academic gains in Ohio’s brick-and-mortar charter schools are large relative to those typically observed in education policy research. Average student performance masks considerable variation, however, as approximately 40 percent of charters post significantly lower annual gains in at least one subject than nearby district schools. Moreover, charters disproportionately serve economically disadvantaged urban students with academic achievement and attendance rates well below Ohio averages. The charter school advantages documented in this report are not sufficient to overcome the substantial academic headwinds these students face.
Nevertheless, the annual academic gains of the average charter school student are comparable to those produced by some of the most effective—and often far more costly—educational interventions. The policy implications are straightforward. Accountability policies appear to play an important role in sustaining quality in the charter sector as a whole, and targeted investments in effective schools can amplify the impact of these high performers. Together, these policies should enable Ohio’s charter sector to continue expanding high-quality educational opportunities for students, particularly those in communities where strong public-school options are limited.
Background
A. Brick-and-Mortar Charter Schools
Ohio law authorized the establishment of charter schools in 1997 to provide additional public schooling options to families trapped in poorly performing school districts. These charter schools (called “community schools” in state law) are tuition-free, open to all Ohio students, and independent of school districts. This independence provides charters some freedom to innovate but, in exchange, they are held accountable for outcomes. School authorizers (or “sponsors”) and state law hold charters formally accountable for their performance. And, because public funding follows students, charters must satisfy the demands of parents to remain financially viable. By expanding educational options for parents, lawmakers created incentives for all public schools—both charter and district—to meet the educational needs of children.[8]
During the 2024–25 school year there were 340 charter schools serving 119,310 Ohio students in grades K–12, approximately 6.3 percent of Ohio’s school-age children and 7.3 percent of its public-school students.[9] This analysis focuses on the 251 site-based, brick-and-mortar charter schools (excluding schools focused on dropout prevention and recovery as well as “virtual” schools) as these schools are most comparable to traditional district schools and were those eligible for the QCSSF program that is the focus of this report.[10] Consistent with policymakers’ intent to provide options to families trapped in struggling districts, Ohio’s charter schools serve primarily low-income and low-achieving students who reside in densely populated urban school districts. As Table 1 indicates, during the 2024–25 school year, the average charter school had 274 students who were predominantly economically disadvantaged (94 percent), Black or Hispanic (69 percent), scored well below average on state tests (between the 28th and 30th percentiles), and struggled with attendance (48 percent missed 10 percent or more of instructional hours).
Table 1 provides a high-level comparison of charter and district schools statewide during the 2024–25 school year. However, because charter schools serve primarily economically disadvantaged children in urban areas, as well as a mix of students who reside outside of Ohio’s largest urban districts (the “Urban Eight”), the birds-eye view in Table 1 does not properly portray the relative funding and performance of charter and district schools. The subsections below provide a more nuanced account of how charters have performed over time and some factors that likely influenced their trajectory.
Table 1. Characteristics of charter schools that are the focus of this study in 2024–25
B. Charter School Performance, 2007–2025
Assessing whether charter schools provide parents with quality educational options, relative to other public schools serving children in their area, requires a statistical methodology that adjusts for the differences in students across sectors. Researchers have converged on a statistical method that does so reliably: schools’ test-score value-added. It captures how much higher or lower student achievement is than one would have anticipated given students’ achievement trajectories in prior school years.[11] In other words, this method assesses the “gains” that schools provide relative to other schools. That’s what I did in this study: estimated statistical models that compare the annual test-score gains of charter schools to the gains of district schools serving similar students who reside in the same school districts.[12]
Figure 1 below provides relative gain estimates across charters (weighted by each school’s enrollment) for grades 4–8, as tests were administered most consistently in those grades.[13] It reveals a steep upward trajectory of charter school performance. Charter students’ average annual gains on English and math tests (the black line) were 0.05 standard deviations below that of nearby district schools during the 2006–07 school year (labeled “2007”). By the 2024–25 school year, however, charter school students outpaced those in district schools by 0.06 standard deviations, on average. These are annual achievement gains. They compound as students progress through school. The estimate for the 2024–25 school year implies that, if a student attended a charter school from grade 4 to grade 8, by grade 8 that student’s test scores would be 0.3 standard deviations higher than they would have been had they attended a nearby district school. This estimated difference is roughly equivalent to charter students getting an additional 75 percent of a school year (130 days of additional instruction) over those five years.[14]
Figure 1. Difference in annual test-score gains between charters and district schools
Though few students attend a consistently high-performing school for five years, the cumulative effects are likely still large for the average student. Even if the cumulative effects were 0.2 standard deviations (two-thirds of the implied cumulative achievement advantage in charter schools today) such effects would be large compared to those of other education interventions.[15] With this benchmark in mind, the improvement of 0.11 standard deviations in average annual test-score gains between 2006–07 and 2024–25 is enormous. Quality varies by individual school, however. For example, during the 2024–25 school year, 26 percent of charters posted average achievement gains in English that were statistically significantly lower than those of nearby district-run schools, and 40 percent of charters posted lower average gains in math.[16]
Note that the historical estimates of charter school performance are like those found in other rigorous studies that assess Ohio’s charter sector, demonstrating steady improvement since 2006–07, a peak in performance in 2018–19, and a steep decline during the pandemic.[17] This analysis is the most recent and, thus, the first to document a complete rebound from 2021–22 to 2024–25. The average charter school is, once again, performing at its 2018–19 peak. Interestingly, achievement gains in math are now at their highest ever.
C. Charter School Accountability, 2007–2025
Notable changes in charter school accountability have contributed to the steep improvements in their performance over time. First, Ohio’s automatic charter school closure law, enacted in 2006, requires poorly performing charter schools—determined in large part based on students’ annual test-score gains—to shut down. This law eliminated schools with low achievement gains (which mechanically contributes to the positive trend in Figure 1) and significantly improved the achievement of students who were displaced by school closure between 2008–09 and 2012–13.[18]
Figure 2 (below) illustrates the force of the closure law by comparing the closure rates of charter schools that had below-average test-score gains in the prior school year (blue line) and those that had above-average value-added estimates in the prior school year (green line).[19] The rate of school closure (for any reason, not just the closure law) for below-average schools quickly climbed to 11 percent by 2010, significantly outpacing the closure rate for schools with above-average test-score gains, which peaked at 4 percent during the 2007–2013 period. This gap in closure rates is remarkable when one considers that the identification of schools as being below- or above-average is a rough approximation.[20] The gap is often higher if this comparison is limited to only those schools whose performance we can categorize with high statistical certainty—those that are well above or well below average. The above/below distinction used in Figure 2 is rough, yet the law’s effects are so pronounced that they nonetheless shine through.[21]
Figure 2. Closure probability of charter schools with above-average and below-average annual test-score gains as compared to all public schools in students’ districts of residence
In addition to the school closure law, other charter school regulations increased steadily from 2007 through 2025. Most notably, the Ohio General Assembly enacted a package of reforms in 2015 that instituted a more rigorous accountability system for charter school sponsors—nonprofit and government organizations that authorize and oversee Ohio charter schools—and imposed a variety of other rules on their financial management and operators. The improvements in charter performance between 2014–15 and 2018–19—despite the diminished link between closure and school performance in that period—were likely at least partly attributable to the 2015 regulatory reforms.
By 2025, after a reduction in closures following the pandemic, the differential closure rates reemerged between charter schools with above-average gains and those with below-average gains. Seven percent of schools (10 schools) with gains below the district average in 2023–24 were no longer in operation during the 2024–25 school year, whereas none of the charters with above-average gains shut down. Switching the comparison to charters with gains above or below the average for nearby district-run schools (as in Figure 1), during the 2023–24 school year 96 of the 226 charter schools for which gains could be calculated demonstrated statistically significantly lower gains than nearby district schools in at least one subject, and 10 of these schools were no longer in operation during the 2024–25 school year. On the other hand, 130 of the 226 charters posted higher annual achievement gains than nearby district schools in at least one subject in 2023–24, and all were still in operation the following year.
D. Charter School Funding, 2020–2025
Recent funding reforms may also help explain the dramatic post-pandemic turnaround, particularly in terms of math achievement. Ohio charter schools have always spent less per pupil, on average, than district schools, but the resource disparity between charter and district schools has been difficult to quantify due to data limitations and differences in how charter and district schools operate. The analysis in this report employs fine-grained school-level spending data and student-level attendance data to more precisely characterize the spending gap.
Since the 2018–19 school year, Ohio’s school-level spending data have included estimates of district schools’ share of central-office expenditures, in addition to their direct operating expenses, making school-level spending data more comparable between charters and district schools.[22] Comparing charter and district spending remains challenging, however, as school-level spending depends on the characteristics of the students a school serves (such as their grade level and special-education and English-language status, as well as the income and property values of district households[23]) as well as the financial cost of operating a school in a particular locale (such as the cost of human and physical capital). To partially address these challenges, this analysis links school-level funding to student-level data, comparing the average expenditures of the charter and district schools attended by students who reside in the same district and who are in the same grade.[24] These spending comparisons are also restricted to students in grades 4–10, so that they correspond to the analysis of student outcomes later in this report.
Figure 3. Spending per pupil among charter schools and nearby district schools serving students in the same district of residence
Figure 3 presents the estimated differences in overall spending per pupil. It reveals that for the average charter student in the sample, nearby district schools outspent their charter school by approximately $4,500 during the 2018–19 school year ($18,091 to $13,503 in 2025 dollars). This difference increased to approximately $5,000 by the 2021–22 school year ($20,042 to $15,044 in 2025 dollars). But the gap subsequently narrowed as the state increased support, primarily via the Quality Community School Support Fund (QCSSF) program. The subset of charter schools that received this funding were able to cut the expenditure gap in half—from $5,000 to $2,500—by the end of the 2024–25 school year, whereas charters that did not receive the funding still had a gap of $5,000 relative to nearby district schools (district schools spent $20,202 per pupil whereas non-QCSSF charters spent $15,275). The QCSSF program is the focus of the remainder of this report.
Quality Community School Support Fund
Governor DeWine and the Ohio General Assembly enacted the 
Quality Community School Support Fund in 2019.[25] Starting in January 2020, high-quality schools that served primarily low-income students could receive up to $1,750 per economically disadvantaged student and up to $1,000 per non-economically disadvantaged student.[26] Lawmakers increased maximum payment amounts significantly beginning in the 2023–24 school year, to $3,000 per economically disadvantaged student and $2,250 for others. Approximately 25 percent of charter schools qualified and, because over 94 percent of students in these schools were economically disadvantaged, the average QCSSF payment more than doubled to $2,957 per pupil.
Prior work examined the impact of QCSSF between 2019–20 and 2022–23, showing that the program benefited students.[27] Here, the analysis focuses on the substantial change in funding starting in the 2023–24 school year. Because the QCSSF program selects schools based on educational quality, it may improve the performance of Ohio’s charter sector by 1) lowering the risk that high-quality schools close for financial reasons and 2) boosting further the performance of these schools by enabling investment in operations and capital. In that sense, it combines the accountability and funding dynamics described above. The analysis below estimates just how much more effective QCSSF schools are, describes how their operations differ in terms of spending and staffing, and, finally, assesses the extent to which higher funding (as opposed to the selection of schools based on quality) caused these differences in performance and operations between QCSSF and non-QCSSF schools.
A. Causal Impact of Attending a Charter School Receiving QCSSF Funding
Ninety-two percent of charter schools (58 of 63) that qualified for the increased funding during the 2023–24 school year did so because they were deemed highly performing according to the QCSSF program’s Criterion 1 (see Table B1 in Appendix B). This criterion requires that a school primarily serve economically disadvantaged students, that its sponsor’s rating is “exemplary” or “effective,” that it have a performance index score (average student achievement on state tests) higher than the district in which it is located in each of the two most recent state report cards, and that it have a “progress” rating of four or five stars on the most recent state report card. This last requirement means that a school’s students need to post annual test-score gains significantly above the Ohio average for similar students.
The analysis below estimates the overall academic benefits for students attending schools that received QCSSF funding during the 2023–24 and 2024–25 school years. Specifically, it estimates the causal impact of attending these schools on students’ achievement and attendance by comparing the academic trajectories of charter school students who did and did not attend schools that received payments. To increase the similarity of students who did and did not attend schools receiving QCSSF funding, the analysis effectively compares charter students who attended the same school at baseline but only some of whom would go on to attend a charter school that received QCSSF funding.[28]
Figure 4. Improvements in student achievement at QCSSF schools
Figure 4 presents the estimated impact of a student attending a school receiving QCSSF payments in 2023–24 and 2024–25 on their achievement in math and English in grades 4–8 and high school (primarily grade 9 for math and grade 10 for English). It indicates that students attending QCSSF schools in those years experienced an immediate increase in math and English achievement of 0.31 of a standard deviation and 0.18 of a standard deviation, respectively. Again, effect sizes of 0.2 or above can reasonably be characterized as “large” compared to other education interventions.[29] These effects occur in the first year or two and are roughly equivalent to students getting an extra 60–75 percent of a school year (100 to 130 days) of additional instruction relative to their charter-school peers who did not attend a QCSSF school.[30]
Figure 5. Reductions in student absences at QCSSF schools
Figure 5 presents the estimated impact of attending a QCSSF school on absence rates (the percent of instructional hours that students missed during the school year) and on rates of chronic absenteeism (the proportion of students who missed 10 percent or more of instructional hours). The results indicate that QCSSF students’ absence rates in grades 4–10 declined by 1.78 percentage points relative to the absence rates of peers who did not attend QCSSF schools. Consistent with this result, rates of chronic absenteeism went down 2.61 percentage points. These effects appear modest given the baseline rates of overall absence (12 percent) and chronic absence (46 percent), but these overall results mask pronounced effects for students in later grades. Focusing on students in grades 8–10 reveals somewhat greater reductions in absence rates (a reduction of 2.42 percentage points) and a large reduction in rates of chronic absence (by 10.20 percentage points). Thus, when it comes to attendance, the primary benefit of attending a QCSSF charter school is a reduction in chronic absenteeism among high school students.[31]
B. Operational Differences between QCSSF and non-QCSSF Schools
Using the same research design and student sample used in the analysis of academic outcomes, one can assess how school operations differed for students who did and did not attend QCSSF schools. Figure 6 (below) reveals that, predictably, the QCSSF schools that students attended reported spending $2,848 more per pupil (in 2025 dollars) than the non-QCSSF charters that their peers attended. The figure also reveals that 58 percent of this spending was directed toward the classroom (such as teachers, aides, counselors, and instructional support staff) and 29 percent was spent on non-classroom administration and operations. The remainder was spent on nonoperating items such as capital outlays, equipment, and interest on debt.
Figure 6. Spending differences students experienced in QCSSF schools
Figure 7. Spending differences students experienced in QCSSF schools, in more detail
The Ohio Department of Education and Workforce further disaggregates classroom and non-classroom expenditures into the categories in Figure 7. Ninety percent of additional classroom expenditures ($1,540 per student) were directed toward instruction (activities involving the interaction of students and teachers or instructional aides and materials like textbooks), but there were also increases of $99 per student in pupil support services, such as counseling, health and psychological services, and social work, and $10 per student in instructional staff support services such as professional development. For a school with 400 students (roughly the baseline average for the subset of charters used in this portion of the analysis), the totals are approximately an extra $615,000 spent on instruction, $40,000 on pupil support, and $4,000 on staff support. The additional nonclassroom expenditures primarily went toward school administration ($383 per pupil) and school maintenance ($354 per pupil)—approximately $150,000 and $140,000, respectively, for a school with 400 students.[32]
If a QCSSF school were to direct all additional instructional dollars toward teachers, as opposed to aides or instructional materials, the increases in instructional expenditures are enough to hire an additional dozen teachers or, instead, give all teachers in the average school a raise of approximately $20,000. The analysis of staffing data—reported in Table 2—indicates that QCSSF schools direct money toward salaries, as opposed to more teachers. Unfortunately, the salary data are messy (e.g., they contain missing and implausible values) and it is unclear how much of the extra instructional spending went toward salaries, as opposed to instructional aides or materials. The analysis indicates that students in QCSSF schools were exposed to teachers who make an additional $15,446 annually. However, ODEW recommends that teacher salaries be divided by teacher full-time-equivalent (FTE) assignments, to account for how much a teacher’s time is spent in a school. Doing so reduces the salary difference from $15,446 to $2,272.
Table 2. Teacher staffing differences in QCSSF schools (2023–24 and 2024–25)
Another challenge with the staffing analysis reported in Table 2 is that the QCSSF schools students attend are much smaller than those of their peers in the sample. Indeed, there are approximately 300 fewer students in the average QCSSF school than in the non-QCSSF schools that comparison students attend (see Table D3 in Appendix D). This may contribute to volatility in statistics scaled by teachers and student counts and, in turn, makes it more difficult to identify precise differences in staffing. That may partially explain the lack of statistically significant differences in student-teacher ratios, for example.[33]
There are some clear takeaways from the staffing analysis in Table 2, however. Students who ended up in QCSSF schools—those who experienced the large academic gains detailed above—were exposed to teachers with higher salaries (between $2,272 and $15,446 more), who turn over less (by 5.2 percentage points, a 14 percent reduction in first-year teachers relative to the baseline of 36.4 percent), and who miss less school (by 1.5 percentage points—a 34 percent reduction in missed school days relative to the baseline of 4.4 percent).[34] But students in QCSSF schools did not have teachers with more advanced degrees or who were more experienced, and the evidence suggests comparable student-teacher ratios.
C. Causal Impact of QCSSF Charter School Funding
The analysis above reveals that charter students who attended QCSSF schools experienced large academic gains compared to their charter-school peers who did not attend QCSSF schools. As anticipated, they also experienced differences in overall expenditures per pupil ($2,848), which their schools spent on instruction and, to a lesser extent, administration and plant maintenance. What the analysis above cannot say, however, is whether the differences in operations and academic effectiveness were because of the extra funding or merely reflect preexisting differences between QCSSF and non-QCSSF charter schools. After all, the QCSSF program provides additional funds to schools that posted student achievement gains above the Ohio average, perhaps because they have more effective curriculum and leadership or stronger school cultures.
To get a sense for how much one can attribute to the extra funding, Table 3 (below) reports the results of a regression discontinuity (RD) analysis that compares the operations and academic outcomes of charter schools that narrowly qualified for up to $3,000 per pupil in extra funding during the 2023–24 year to those of nearly identical charter schools that narrowly failed to qualify for the extra funding in that year.[35] A strength of this design is that it can identify the causal impact of funding on schools near the performance cutoff for the QCSSF program—which were of average to above-average quality. A weakness of the design is that the results may not be generalizable to more poorly or highly performing charter schools. Additionally, the design yields less precise estimates than the difference-in-differences design presented above. Thus, the results should be considered rough estimates of the causal impact of receiving greater state funding on the operations and academic effectiveness of charter schools.
The results in Table 3 suggest that approximately 50 percent of the achievement effects of attending QCSSF schools were due to their having greater funding, and that 100 percent of the reductions in absence rates were due to the extra funding. They also suggest that the overall differences in spending, as well as differences in classroom and non-classroom spending, were fully attributable to additional funding provided by the QCSSF program.[36] Finally, the estimated effects of QCSSF on teacher turnover and absences are largely similar, but the estimated impacts on teacher salaries ($3,873 and $4,146, depending on the measure) are on the lower end of the range presented above ($2,272 and $15,446, respectively). Given these results, it seems reasonable to state that QCSSF funding is responsible for approximately a $4,146 increase in annual teacher salaries.
Table 3. Percent of impacts plausibly due to QCSSF funding in 2023–24 and 2024–25
The RD analysis makes clear that much of the effectiveness of QCSSF schools is due to the increase in funding and not merely because the program selects schools based on their performance. The RD analysis has some important limitations, however. The design does not perfectly distinguish between schools that did and did not get funding and, thus, the magnitudes of the estimated coefficients are likely somewhat muted.[37] The RD analysis also focuses on the impact among schools that were above average (or nearly above average) in terms of their annual student achievement gains. Thus, one cannot be sure that the estimated effects would have been the same for the very best and very worst charter schools. Finally, we cannot completely rule out that the academic outcomes are driven by motivated students selecting into these schools when the QCSSF award occurred. Although the analysis examines student-level gains in achievement—thus holding fixed student achievement at baseline—the students who experienced academic gains in schools that received QCSSF funding were more mobile than comparison students in schools that did not receive QCSSF funding. These dynamics present the possibility that something other than instructional inputs drove improvements in schools that received QCSSF funding.
Summary and Implications
The performance of Ohio’s charter schools has improved dramatically over the past two decades. In the 2006–07 school year, charters on average produced lower academic gains than nearby district schools serving similar students. Today, the typical charter school generates significantly larger achievement gains than district schools serving the same student populations—despite operating with substantially less funding. This analysis links the remarkable turnaround to public policy.
Ohio’s automatic charter school closure law, sponsor evaluation system, and other quality-control efforts have increased the consequences for persistently weak performance, while state initiatives that provide additional funding to effective charter schools have rewarded strong results. The above analysis suggests that these forces have gradually reshaped the sector by allowing more effective schools to expand while removing those that consistently underperform. Recent increases in charter school funding have also further enhanced the performance of high-quality schools serving economically disadvantaged students, who stand to benefit most from additional resources. Although these charter schools still spend approximately $2,500 less per pupil than nearby district schools and did not begin receiving the larger payments until January 2024, student achievement and attendance nonetheless increased significantly during the 2023–24 and 2024–25 school years—far beyond the academic gains in neighboring district schools and non-qualifying charter schools. It is possible that the academic benefits of these investments will grow even larger in subsequent years as schools have greater opportunity to deploy the resources strategically.
Some observers may question whether the improvements documented in this report simply reflect changes in the students who charter schools serve rather than genuine improvements in school performance. The analysis addresses this concern directly. The estimates compare the academic growth of charter school students with that of district school students who live in the same communities and have similar prior achievement levels. As a result, they capture differences in the gains students make after enrolling in charter schools rather than differences in the characteristics of students who choose them. The findings are also consistent with prior research documenting steady improvements in Ohio’s charter sector over time. Taken together, the evidence suggests that the sector’s rising performance reflects real improvements in school quality rather than shifts in student composition or other statistical artifacts.
The results carry important implications for policymakers. Ohio’s experience suggests that strong accountability policies can play a central role in improving the quality of charter schools over time. By imposing consequences for persistently weak performance, the state’s closure law and sponsor evaluation system have helped remove ineffective schools while allowing stronger schools to expand. At the same time, targeted funding increases for high-performing charter schools have further improved outcomes. In combination, these policies create a system that both rewards effectiveness and provides successful schools with the resources needed to serve students well. The results of this report suggest that maintaining this balance—rigorous accountability paired with strategic investment in effective schools—is a promising strategy if Ohio hopes to build upon charter schools’ considerable progress.
Acknowledgments
We wish to thank Dr. Stéphane Lavertu for his fine work conducting the analysis and authoring this report. Special thanks to the Ohio Department of Education and Workforce for providing the data necessary for the analysis. On the Fordham team, we appreciate the thoughtful feedback from our colleagues Michael J. Petrilli and Chester E. Finn, Jr., as well as assistance from Jeff Murray. Thanks also to Andy Kittles for laying out the publication.
—Aaron Churchill and Chad L. Aldis
Thomas B. Fordham Institute
Endnotes
[1] Center for Research on Education Outcomes, As A Matter of Fact, National Charter School Study III (2023).
[2] Tom Kane, “The cost of the charter school cap,” CommonWealth (October 5, 2016), and Jason L. Riley, “New York’s charter schools live up to their promise,” Wall Street Journal (February 3, 2026).
[3] For additional policy detail, see Aaron Churchill, Reinventing Ohio’s Charter School Sector, 2015–2023, Thomas B. Fordham Institute (2023).
[4] In Ohio legalese, charter schools are known as “community schools.”
[5] Aaron Churchill, “Has Ohio erased its charter funding gaps?” Thomas B. Fordham Institute (January 21, 2025).
[6] Stéphane Lavertu, The Impact of Increasing Funding for High-Performing Ohio Charter Schools, Thomas B. Fordham Institute (2024).
[7] The analysis includes all site-based, brick-and-mortar charter schools, including “special education” charters in which a majority of students have independent education plans.
[8] Ohio lawmakers have also expanded school options by providing families vouchers to send their children to private schools. This study, however, focuses on charter schools, which are public school options and do not require a voucher.
[9] These statistics are unlikely to match official statistics perfectly, as they are based on the author’s calculations using student-level data and include students in K–12 only.
[10] There are also seven online “virtual” schools and 82 schools that focus on dropout prevention and recovery that are legally charter schools, but they are not comparable to traditional public schools.
[11] For an accessible review, see Cory Koedel, “Value-Added and Growth Models in Education,” in Live Handbook of Education Policy Research, ed. Douglas Harris (Association for Education Finance and Policy, 2025), accessed December 2, 2025. For a technical review, see Andrew Bacher-Hicks and Cory Koedel, “Estimation and Interpretation of Teacher Value Added in Research Applications,” in Handbook of the Economics of Education, vol. 6, ed. Eric A. Hanushek, Stephen Machin, and Ludger Woessmann (Amsterdam: Elsevier, 2023), 93–134.
[12] See Appendix C, section titled School Value-added #1, for details of the statistical procedure.
[13] These estimates are distinct from but comparable to Ohio’s official school value-added estimates.
[14] Research indicates that students typically experience annual test-score growth of 0.314 of a standard deviation in reading and 0.422 of a standard deviation in math in grades 4–8. Dividing these gain estimates by the typical growth rates yields a fraction of a school year, which can be multiplied by 180 (the typical number of instructional days in a school year) to estimate additional days of learning. See Carolyn J. Hill, Howard S. Bloom, Alison R. Black, and Mark W. Lipsey, “Empirical Benchmarks for Interpreting Effect Sizes in Research,” Child Development Perspectives 2, no. 3 (2008): 172–77.
[15] Matthew A. Kraft, “Interpreting Effect Sizes of Education Interventions,” Educational Researcher 49, no. 4 (2020): 241–253.
[16] Specifically, the difference between a charter school’s gain estimate and the gain estimate among nearby district schools was negative and statistically significant at the 5 percent level.
[17] These estimates are comparable to those found in past Fordham Institute reports as well as to CREDO’s estimates for brick-and-mortar charter schools. See Stéphane Lavertu, The Impact of Ohio Charter Schools on Student Outcomes, 2016–19 (Columbus, OH: Thomas B. Fordham Institute, 2020); and Stéphane Lavertu, Ohio Charter Schools After the Pandemic: Are Their Students Still Learning More Than They Would in District Schools? (Columbus, OH: Thomas B. Fordham Institute, 2024). For 2018–19 estimates using a credibly causal quasi-experimental design, see Stéphane Lavertu and Long Tran, “For-Profit Milk in Nonprofit Cartons? The Case of Nonprofit Charter Schools Subcontracting with For-Profit Education Management Organizations,” Journal of Policy Analysis and Management 44, no. 1 (2025): 236–265. For CREDO’s estimates, see Center for Research on Education Outcomes (CREDO), Charter School Performance in Ohio (Stanford, CA: Stanford University, 2014); and Margaret E. Raymond et al., As a Matter of Fact: The National Charter School Study III 2023 (Stanford, CA: Center for Research on Education Outcomes, 2023). As explained elsewhere, CREDO’s main statewide estimates can be misleading because they include schools focused on dropout prevention and recovery as well as virtual schools, which serve very different student populations; see Stéphane Lavertu, “CREDO Got It Wrong: Ohio’s Charter Sector Is Strong and Getting Better,” Thomas B. Fordham Institute (commentary).
[18] Deven Carlson and Stéphane Lavertu, “Charter School Closure and Student Achievement: Evidence from Ohio,” Journal of Urban Economics 95 (2016): 31–48.
[19] These value-added estimates once again capture the achievement gains of a school’s average student relative to gains for similar students in the residential district, but without differencing out the estimated value-added of nearby district schools. The procedure for generating these estimates is described in Appendix C, in the section titled School Value-added #2. The results are similar when the difference in gains is used to identify high and low performers.
[20] Schools are categorized based on a single year of achievement gains and many schools categorized as above or below are very close to the average gain of zero. Consequently, the differences between above- and below-average schools often fail to reach statistical significance.
[21] The law identified schools for closure based on multiple prior years of performance data, starting with 2006–07 and 2007–08 gains estimates. Thus, increases in closure should occur beginning in 2009, which is consistent with the trend in Figure 2. The figure also reveals a temporary narrowing of the gap in closure rates between 2013–14 and 2017–18 and between 2021–22 and 2022–23. These periods line up with the increased noisiness of the gains estimates and, most importantly, the temporary provision of “safe harbor” from the consequences of the closure law, due to the introduction of PARCC assessments in 2015 (lasting between 2014–15 and 2016–17) and, later, the onset of the COVID-19 pandemic in 2020 (lasting from 2019–20 to 2021–22). Note that as of 2024–25, the gap in closure rates between schools with above-average test-score gains (0 probability of closure) and below-average gains (0.07 probability of closure) has reemerged and is the highest since the 2008–09 school year. The closure law is back in full force and high-performing charters now benefit from significantly greater funding that should help them withstand financial shocks and remain open (more on that below). None of the 2023–2025 closures are formally attributed to the closure law, but the law being in force creates incentives and might lead to anticipatory effects such that schools close on their own before the law forces them.
[22] Districts’ central offices provide services to their schools but, until 2018–19, the financial cost of these services was reflected only in district-level expenditure data. Data for the funding analysis were obtained from the Ohio Department of Education and Workforce’s Report Portal: https://reports.education.ohio.gov/overview (accessed December 20, 2025).
[23] Property values affect local district tax revenues, and charter schools do not have access to that revenue (except for a limited number of charters in Cleveland). State funding serves as a partial counterweight to disparities in local tax revenues, allocating more funds to both charter and district schools that serve students who are more expensive to educate.
[24] Specifically, average differences in school spending were calculated by estimating a regression model that estimates, for each charter school, the difference between their spending and the spending of district schools while holding constant student grade and district of residence. The sample of students used in the analysis is the grade 4-10 sample of students used in the calculation of value-added for the 2018–19 through 2024–25 school years. The estimated differences were then combined with average spending in those schools to generate a prediction of average spending per pupil (weighted by charter enrollments) in charter students’ “nearby district schools.”
[25] See Appendix B for statistics on QCSSF payments and schools that received the funding. For detailed information on the program’s first few years, from 2020 through 2023, see Stéphane Lavertu, The Impact of Increasing Funding for High-Performing Ohio Charter Schools: The Quality Community School Support Fund, 2019–23 (Columbus, OH: Thomas B. Fordham Institute, 2024).
[26] Due to an appropriation cap, the maximum per-pupil amounts were not awarded through to the 2022–23 school year. In that year, for instance, 45 percent of charter schools qualified and received an average payment of $1,367 ($1,442 in 2025 dollars). Figure 3 seems to indicate that charters getting payments between 2019–20 and 2022–23 did not in fact spend that money, but that is not the case. Schools that narrowly qualified for payments in those years did indeed spend more than those that narrowly failed to qualify (see Lavertu, 2024). But, between 2018–19 and 2022–23, the average charter school that did not qualify for the program spent just as much per pupil as the average school that qualified. It may be that, in the absence of the program, non-qualifying schools would have spent approximately $1,500 more per pupil, on average, than qualifying schools. Indeed, non-qualifying schools were more likely to be in relatively affluent and highly performing districts. And non-qualifying schools include some with exceptionally high spending levels, such as schools that focus on students with autism. Finally, it could also be that non-qualifying schools were more likely to spend pandemic funds or raise other revenue during this period.
[27] Stéphane Lavertu, The Impact of Increasing Funding for High-Performing Ohio Charter Schools: The Quality Community School Support Fund, 2019–23 (Columbus, OH: Thomas B. Fordham Institute, 2024).
[28] Appendix D describes this “difference-in-differences” design and demonstrates that the students suddenly exposed to QCSSF funding had prior trends in achievement and school characteristics that were similar to comparison students who would not go on to attend schools that received QCSSF payments. It also describes the statistical approach used to isolate the impact of the program during the 2023–24 and 2024–25 school years, as schools that received funding in those years sometimes did so by qualifying for the lower payment amounts
during the 2022–23 school year.
[29] Matthew A. Kraft, “Interpreting Effect Sizes of Education Interventions,” Educational Researcher 49, no. 4 (2020): 241–253.
[30] Research indicates that students typically experience annual test-score growth of 0.314 of a standard deviation in reading and 0.422 of a standard deviation in math in grades 4–8. Dividing these gain estimates by the typical growth rates yields a fraction of a school year, which can be multiplied by 180 (the typical number of instructional days in a school year) to estimate additional days of learning. See Carolyn J. Hill, Howard S. Bloom, Alison R. Black, and Mark W. Lipsey, “Empirical Benchmarks for Interpreting Effect Sizes in Research,” Child Development Perspectives 2, no. 3 (2008): 172–77.
[31] There are no attendance benefits in elementary and middle-school grades (see Table D3 of Appendix D).
[32] For a school with 300 students (approximately the average for Ohio charters) the totals equate to an extra $460,000 spent on instruction, $30,000 on pupil support, and $3,000 on staff support. The additional non-classroom expenditures for school administration and maintenance were approximately $115,000 and $106,000, respectively.
[33] This difference in enrollments is large and reflects the unique sample of schools that enable the difference-in-differences analysis. Fortunately, the regression-discontinuity analysis below focuses on schools with similar enrollments and yields comparable estimated effects.
[34] The reduction in the percent of teachers in their last year is nearly identical to the reduction in teachers in their first year, but that measure is available for one fewer year so only the percentage of teachers in their first year is reported.
[35] See Appendix E for a description and empirical validation of the research design, as well as tabular results across all school-level outcomes.
[36] Estimates for sub-categories of spending are typically too imprecise to yield statistically significant results, but the estimated coefficients for these sub-categories are generally similar. The main difference is that the regression-discontinuity analysis suggests larger increases in transportation spending (approximately $368 per pupil). See Table E3 in Appendix E.
[37] For example, some schools that narrowly missed out on funding in 2023–24 qualified for funding in 2024–25, and as Appendix E describes, the “running variable” is based on the academic criteria of Criterion 1 and does not account for the few schools that may or may not have qualified based on other factors (e.g., the proportion of students who are economically disadvantaged).
