Charting a Path Forward for Charter Schools
A Policy Brief by the Committee for Economic Development of The Conference Board
February 13, 2018
Over the past decade, charter schools have gained a prominent role in the debate over how to improve K-12 education, becoming one of those rare ideas that attracts bipartisan support. Yet as charter schools have expanded and begun educating more American school children, so has discussion over the role charter schools should ultimately play in our public school system. This policy brief provides an overview of what charter schools are (and are not) and what we know about their effectiveness, with the aim of facilitating more knowledgeable discussion among policymakers and other stakeholders.
- What Are Charter Schools?
- What Proponents Say about Charter Schools
- What Critics Say about Charter Schools
- How Charter Schools Operate
- Effectiveness of Charter Schools
- Recommendations for Future Charter School Policy
What Are Charter Schools?
Charter schools are publicly funded schools that typically are run by a group or organization other than a local school district.1 These schools operate independently under a charter (or contract) with the district, state, or other authorized entity.2 Broadly speaking, charter schools operate as an extension of “traditional” public schools run by local school districts. In contrast with the traditional public school experience in which students typically attend a school assigned to them based on where they live, charter schools are schools of choice, and parents must apply to enroll their children in a charter school.3,4
State (and local) laws governing the charter school sector vary greatly, but in general, charter schools are exempt from many of the state and local rules and regulations that govern other public schools.5 This allows charter schools to be innovative in their approach to education and gives them greater managerial flexibility. Charter schools typically set their own curriculum, hire staff outside the school districts’ own processes, and control other aspects of school operation such as the length of the school year and discipline policies.6 Charter schools often have enrollment caps (restricting the number of students enrolled at each grade level) and are often allowed to develop their own “backfilling” policies, meaning that even if a school has open seats, it can opt not to accept midyear enrollments or new students entering higher grade levels.7 In exchange for this increased autonomy, charter schools are, at least in theory, held more accountable for improving the academic performance of the students who attend them.8
In keeping with their status as publicly funded schools, charter schools are nonsectarian, may not charge tuition, and must comply with the same federal civil rights laws and regulations as other public schools.9 Generally, charter schools must be open to all students who wish to enroll, except when schools reach their enrollment capacity. Most states require that oversubscribed charter schools hold admission lotteries, with the intent of giving every new applicant an equal chance at acceptance. Although charter schools are exempt from some of the rules governing traditional public schools, most are still accountable to a public authority via their charter. In many cases, the local school district where the charter school is located is the governmental authority responsible for determining whether the charter school has met its performance obligations under its charter.
The motivation behind charter schools
Minnesota passed the first charter statute in 1991 and subsequently opened the first charter school in 1992.10 Legislators in this early innovator state were inspired by the work of Ray Budde, an educator who first developed the idea of educational charters in 1974.11 The concept did not gain traction, however, until more than a decade later, when reports such as A Nation at Risk, issued by the National Commission on Excellence in Education in 1983, framed the state of American public education as a national crisis. This helped set the stage for greater public acceptance of restructuring public schools as a possible path to improving outcomes.12
Private or Public?
One common question is whether charter schools are public or private schools.
Traditional public schools are funded, owned, and operated by local and state governments, while private schools are owned and operated by private organizations, and draw primarily on private funding sources such as tuition, donations, and endowments. The typical charter school straddles this divide; it is funded by taxpayers and overseen by local and state governments, but is owned and operated by a private organization or group. For example, a local education agency (LEA) might be responsible for ensuring a charter school fulfills its charter obligations, but the LEA does not dictate how the school will accomplish those goals (as long as it complies with state charter laws).
To clear up confusion, the US Department of Education issued a statement recognizing that the US government classifies charter schools as public schools.* The phrase public charter school is therefore redundant, although it is still commonly used by advocates and legislatures alike.* US Department of Education, Office for Civil Rights, Dear Colleague Letter, May 14, 2014.
In 1988, Al Shanker, then president of the American Federation of Teachers, publicly endorsed the concept of establishing new schools to improve school quality.13 While today’s charter schools do not exactly mirror Shanker’s vision for “new” schools, the current charter school movement does reflect his initial description of autonomous yet accountable public schools that would be free to try out new ideas for improving student learning and that would foster greater competition through parental school choice.14
Charter schools today
While still a relatively small part of the overall public school system, charter schools have grown substantially since the 1990s, both in terms of the number of schools and enrollment. In 2004, approximately 887,000 students attended charter schools. As of the fall of 2014 that number had tripled to 2.7 million students—or 5.4 percent of US public school students—who attended 6,747 charter schools (see Chart 2).15 By 2016, 43 states plus the District of Columbia had charter school laws in place.16
Charter schools tend to be concentrated in urban areas. Fifty-six percent of charter schools are located in cities compared to only 25 percent of traditional public schools (see Chart 3).17 Some urban districts boast a robust charter sector. Nearly half of Washington, DC’s school-aged population is enrolled in a charter school.18 In more rural states like Kansas, however, charter schools educate less than 1 percent of public school children (see Chart 4).19
Parental demand may be a key factor driving the geographic distribution of charter schools.20 Each school needs a certain minimum number of students to be financially viable, so the density of potential students in an area affects the number of schools the area can support. Factors such as travel time to a school affect parental choice, especially for parents of younger students. The more compact geographies of urban areas contribute to the greater viability of charter schools in these areas. Perceptions that urban public schools are of lower quality may also be a factor driving higher demand for charter schools in urban districts relative to other types of school districts. In a majority of large urban school districts, student test scores fall below the national average, and families living in areas where traditional public schools typically underperform are more likely to seek out alternatives.21
Compared to traditional public schools, charter schools enroll a higher proportion of minority students. Twenty-seven percent of charter school enrollees identify as black (compared to 15 percent at traditional public schools), and 31 percent identify as Hispanic (compared to 25 percent at traditional public schools). Charter schools also are slightly more likely to serve high concentrations of low-income students, with 35 percent of charter schools having a student body where 75 percent or more of the students qualify for free or reduced-price lunch (compared to 24 percent of traditional public schools).22 These differences in student demographics are likely due in part to the concentration of charter schools in urban areas, where traditional public schools also serve a disproportionate number of low-income, minority children.
The number of charter schools nearly doubled in a decade.
Number of Charter Schools, 2004–2014Sources: Institute for Education Science, National Center for Education Statistics (NCES), Digest of Education Statistics, October 2016; NCES, September 2015; NCES, January 2015.
What Proponents Say about Charter Schools
Advocates of charter schools point to four core advantages: autonomy, accountability, choice, and improved performance for all public schools (including noncharters) due to healthy competition arising from choice.
Charter advocates argue that traditional public school systems have a poor track record in turning around underperforming schools and that students suffer because rigid state and local rules prevent schools from making needed changes. Advocates argue that because charter schools face fewer regulations and less “red tape,” they are freer to focus time and resources on the choices and activities that really matter in improving student outcomes.23 Moreover, they contend that thanks to charter schools’ greater latitude to experiment with new educational practices and models, charter schools can serve as innovation sites. Traditional public schools later can adopt practices that were first shown to be effective at charter schools, thus benefiting even noncharter school students.
With regard to accountability, advocates maintain that the requirement that charter schools go through an authorization process before opening (and during periodic reauthorizations) means that charter schools face higher standards of accountability than traditional public schools.
Proponents of charter schools argue that the schools serve students by providing free alternatives to underperforming or underresourced district schools. Even when highquality neighborhood schools are available, charter schools still may appeal to families looking for a learning environment that better fits the particular needs and learning styles of their children.
Advocates point to the fact that because charter schools can offer curricula that focus on specific competencies or particular educational models, access to charter schools gives families greater choice in how their children are educated.
Finally, because charter schools are schools of choice and their viability depends on continued demand for their services, advocates maintain that these schools have a greater incentive to provide a high-quality education. Some charter school advocates see the potential for charter schools to improve the performance of noncharter schools as well, by introducing market forces into the overall public school system. These advocates believe that competitive pressure will incentivize existing traditional schools to improve in order to retain students and their associated funding.
What Critics Say about Charter Schools
Many criticisms of charter schools focus on charter schools’ impact on existing public schools and the students who attend them. Critics of charter schools contend that charter schools negatively impact the public school system by diverting funds from traditional public schools. Citing evidence that the average charter school produces no better outcomes than traditional public schools serving similar students, critics argue that opening charter schools is not the best use of limited public resources, particularly in areas with low population densities.
In areas where the traditional public schools are underperforming, students with more motivated, engaged, or informed parents are most likely to take the initiative to search for and apply to promising charter schools. Critics worry that because enrolling in a charter school requires more parental initiative, the students “left behind” at neighborhood schools disproportionately will be those who are most disadvantaged and most in need of additional educational resources, as their school loses enrollment and funding.
Another common concern is that charter schools’ exemption from traditional public school oversight makes it more likely that poorly managed charter schools or those employing exclusionary practices will go undetected.24 Critics point to cases of mismanagement by charter schools as examples of the danger of creating a parallel, less-regulated system of public education. They argue that policymakers should focus on improving existing traditional public schools rather than supplanting them with new schools.
The number of students attending charter schools tripled between 2004 and 2014.
Fall enrollment in charter schools (in millions) and as a percent of total public school enrollment, 2004–2014Sources: Institute of Education Sciences (IES), National Center for Education Statistics (NCES), Digest of Education Statistics, 2016 Tables and Figures, Table 216.20, October 2016. 2015 Tables and Figures, Table 216.20, September 2015. 2014 Tables and Figures, Table 216.20, January 2015.
How Charter Schools Operate
Because state and local laws governing charter schools vary widely, charter schools also vary considerably across states and, even, within the same state.
Establishing a charter school
As their name indicates, every charter school has a charter, or contract, which is a fixed-term agreement spelling out how the school will operate and what it intends to achieve. Essentially, the charter outlines what the school is promising in return for receiving public funding and greater autonomy. Many charters contain explicit performance goals that the agency responsible for renewing a school’s contract can use to hold the school accountable. The charter is the backbone of the chartering process, but it is also just the beginning of opening a charter school and keeping it open. Once the charter school founders have drafted a charter agreement in accordance with state law, they must seek approval from an authorizing entity.
Fifty-six percent of charter schools are located in cities compared to only 25 percent of traditional public schools.
Percent of charter and traditional public schools, by locale (2014–15 school year)Source: Institute of Education Sciences (IES), National Center for Education Statistics (NCES), Digest of Education Statistics, 2016 Tables and Figures, Table 216.30, October 2016
Charter school authorizers
Every state with a charter school statute also has a process for recognizing charter school authorizers. Authorizers are responsible for reviewing new charter school applications and for approving or denying charter renewals. The majority of authorizers are government agencies, principally local and state education agencies (LEAs and SEAs, respectively).25 Some states have independent charter authorizers or permit existing nongovernment entities, such as institutes of higher education, to act as authorizers.26 If a state does permit multiple authorizer types, their jurisdictions may overlap, meaning that a prospective school will have more than one authorizer from which to choose when submitting an application. Another authorizer configuration is to have a primary and secondary authorizer— for instance, if a state allows both LEAs and the SEA to authorize charter schools, the state might make the statewide authorizer, the SEA, an authorizer of “last resort” that is only available if an applicant wants to appeal the decision of the primary authorizer, the LEA.27
To complement their role in approving and renewing charter contracts, authorizers also provide oversight—via financial audits, onsite visits, and report reviews—to monitor progress and regulatory compliance of schools in their portfolio. Authorizers, as individual entities, vary in their approach and involvement when problems are identified at the schools they oversee.28 State law can also help shape the extent to which authorizers will hold charter schools accountable by specifying how authorizers should operate.29 The greatest oversight power of an authorizer is the ability to issue, renew, or revoke a school’s charter. Failure to meet charter performance standards (including poor academic performance), financial or accounting issues, or regulatory noncompliance can result in revocation of a school’s charter causing it to close.
On average, about a third of charter applications are approved.30 Beyond simple approval and renewal of charters, charter laws can impose other restrictions that limit the number of charter schools. Some states have chosen to legislate caps on the number of charter schools in the state or per existing school district.31 In some jurisdictions only new schools can apply to be a charter school, while other jurisdictions allow existing public schools to reorganize as charter schools.
Authorizing timelines—reflecting how often a school must have its charter reviewed and reapproved—also vary. The National Association of Charter School Authorizers (NACSA), a membership association for charter authorizers, recommends that charters be granted five-year contracts.32 While some states adhere to this benchmark, others have established their own minimum and maximum lengths for granting charters, with a few permitting contracts up to 15 years.33
On average, the charter school sector receives less funding per student than traditional district schools.
Charter school funding
Many charter schools struggle with the same funding issues as traditional public schools— principally, how to achieve better results with the same or fewer resources. Like traditional district schools, taxpayers fund charter schools, which receive a mix of local and state funding that is tied to enrollment.34 This “base-level” funding, the formula for which is outlined in states’ charter statutes, is intended to cover the basic cost of educating a child.
In most states, less than 5 percent of public school students attend charter schools.
Percent of public school students enrolled in charter schools, by state (2014-15 school year)Note: Alabama passed its first charter school law in 2015. Source: Institute of Education Sciences (IES), National Center for Education Statistics (NCES), Digest of Education Statistics, 2016 Tables and Figures, October 2016.
On average, the charter school sector receives less funding per student than traditional district schools.35 One reason for this difference is the variety of formulas used to distribute per-student aid to charter schools. For instance, in some states, charter schools receive funding based on the same funding formula used for traditional public schools while in others, the law stipulates that charter schools must receive funding only up to a certain percentage of the per-pupil funding for traditional public schools. Some charter schools receive only state funds while others receive both state and local funds.36
In addition to this per-student funding, charter schools often are eligible for categorical aid (i.e., aid tied to specific programs, such as special education), or funding tied to initiatives such as reducing classroom size. There is some anecdotal evidence that charter schools do not always know what additional aid is available to them, and therefore, do not apply for it—which may contribute further to funding disparities between charter and traditional public schools. Charter schools also are eligible for federal funding, and in the past, specific federal aid has been set aside to support charter schools.37 Thirty jurisdictions also provide some funding to help charter schools manage the costs associated with acquiring and maintaining school facilities.38 Charter schools also may rely on private funding to supplement the public funding they receive.39
Charter school management
Charter schools must accomplish the same managerial functions—for example, overseeing budgets, curriculum, and staffing—that school districts and their associated school boards handle for traditional public schools.40 How charter schools accomplish these functions, however, depends on how the school chooses to organize itself and on what state law permits.
The organization that holds a school’s charter and the organization that actually operates the school on a day-to-day basis is often the same entity but need not be. Some charter schools are completely independent, meaning they do not contract with any outside source to help run the school. Other approaches to allocating management responsibility are possible. For example, a group of educators could establish a charter school and be awarded the charter contract, but then choose to contract out some or all of the school’s management to a third party. So even if a charter school is started by a nonprofit group that holds the school’s charter, contracts with outside groups for school management or other services can introduce for-profit elements.41
As a result, determining the extent to which a particular charter school is primarily “nonprofit” or “for-profit” can become muddled in some cases. However, in general most charter schools are managed by nonprofit organizations, and the vast majority of charter school students attend nonprofit schools.42 Some state charter laws require that any applicant for a charter must be a nonprofit entity or group.43 While research on the question of the relative effectiveness of nonprofit and for-profit charter schools has been limited, it suggests that charter schools run on a for-profit basis are less effective than their nonprofit counterparts.44
In addition to the nonprofit and for-profit distinction, another important characteristic of charter schools is the extent to which they are independent or part of a larger network. An independent school—also known as a “standalone” or “mom and pop” school—is a charter school created by a group or organization that is not formally connected to a larger group of charters schools. A school founded by community members or teachers is an example of an independent charter school. In contrast, network schools are schools that are affiliated with an 9 Policy Brief organization that operates multiple schools or provides operating guidelines or standards that schools must adhere to in order to be part of the network.45 Some of the most well-known names in the charter sector—such as Knowledge is Power Partnership (KIPP) and Uncommon Schools—are examples of organizations that manage or guide a network of charter schools.46 Some management organizations, such as KIPP, operate nationally, while others remain more regionally focused (such as Success Academies in New York City). Although fully independent charter schools reflect the earliest vision of the charter school movement, network schools are becoming more common today. The National Alliance for Public Charter Schools (NAPCS) estimates that in 2014–15, management organizations ran 40 percent of charter schools, up from about a third in 2010–11. Among network schools, approximately 37 percent are operated on a for-profit basis, or 15 percent of all charter schools overall.47
Potential Benefits of Network Schools and Management Organizations
Being a network school offers potential advantages. Management organizations provide organizational structure and other crucial supports, thereby making it easier to get a school up and running. The brand name associated with a successful management group can help network schools attract more private capital and high-quality teachers. Management organizations also help successful educational models achieve “scale” by providing individual schools a clear path forward on how to launch a school and implement the model. However, management organizations that oversee many schools or schools that are geographically dispersed run the risk that their program design will not achieve the same results in different areas serving students of different backgrounds or educational needs.
Effectiveness of Charter Schools
What policy makers, parents, and educators really want to know is how effective charter schools are when compared to traditional public schools. Gauging the effectiveness of charter schools poses challenges, however. (See “Measuring Effectiveness,” p. 10.) First, effectiveness depends in part on how one defines the goals of our public education system. Almost everyone agrees, though, that core outcomes include what students learn while still in school as well as later outcomes such as college completion and employment.
Charter schools may generate benefits for students into their early adult years.
How student outcomes compare between charter schools and traditional public schools
The last decade has generated a growing number of high-quality studies of the effectiveness of charter schools. In addition, the increasing maturity of the sector means that we can now begin assessing longer-term effects of charter school attendance on students who attended such schools in the 2000s. Just in the last five years, high-quality studies have come out suggesting that charter schools may generate benefits for students into their early adult years.
Table 1 (pp. 12–13) summarizes a set of recent high-quality studies. The list includes studies that were either lottery or matched comparison studies from 2013 on, as well as a meta-analysis of 52 highquality studies. In addition, it includes two CREDO (Center for Research on Education Outcomes at Stanford University) studies: one of charter schools in 27 states that enroll 95 percent of charter students in the US, and the other of Ohio charter schools. CREDO uses a “virtual peers” comparison methodology that has been subject to criticism.
Measuring Effectiveness: Lotteries, Matched Comparison Groups, and Meta-Analyses
Because charter schools are schools of choice, assessing their effectiveness compared to typical public schools poses challenges. For example, educational achievement is strongly correlated with students’ socioeconomic status: more advantaged students tend to score higher on academic achievement tests. Test scores also depend on the student’s previous skills and knowledge, as well as factors such as whether the parents are highly motivated or involved in their child’s education. For these reasons, simply comparing scores of charter school students and students at noncharter schools tells us little. If charter school students score higher (or lower) on average, is it because charter schools are better (or worse) than traditional public schools? Or is it because the students who attend charter schools differ on average—in terms of their socioeconomic status, their prior academic achievement, their parents’ (or their own) level of motivation, or some other factor that affects their outcomes? Is it the school or the students who choose the school that are making the difference?
For these reasons, when assessing the effectiveness of charter schools, researchers give the most credence to studies that use “lottery” methodologies. Many charter schools have more applicants than spaces, and in these cases it is often required that spaces be assigned through a lottery where every applicant has an equal chance of gaining admission. On average, winners and losers of the lottery should be similar in all ways (such as parental motivation and previous academic achievement) except whether they won admission to the charter school. This allows researchers to compare outcomes for those who attend charter schools and those who do not, knowing that the two groups of students are the same on average. Therefore, it is the school, and not differences in the students, that is driving the differences in outcomes.
There is a tradeoff with the lottery studies, however. Because such studies can be conducted only when charter schools are “oversubscribed,” they cannot be done for all charter schools, many of which have unfilled seats. Because oversubscribed schools are the most popular schools among parents, it is likely that the charter schools in lottery studies are disproportionately some of the best charter schools, and therefore not representative of all charter schools in the US. For these reasons, lottery studies tend to include relatively few schools, often in one geographic area. (See the third and fourth columns in Table 1.)
To compensate for these limitations, researchers also make use of “matched comparison” studies. Matched comparison studies create a comparison group of noncharter school students that statistically controls for all the student factors that the researcher is aware of (and has data for) that might affect the outcomes being studied. Such factors include demographic data like socioeconomic status and the student’s previous test scores. These analyses typically must exclude “unobservable” factors, like parental motivation, that are not part of administrative datasets. The researcher then compares outcomes for the charter students and the matched comparison group. This weakens confidence in the findings to some extent (the charter students and the comparison group may differ in important ways) but does allow researchers to study the effectiveness of larger numbers of more typical charter schools.
Finally, the strongest source of evidence is a meta-analysis, which combines the results of multiple studies with strong methodologies (for charter schools, lottery or matched comparison studies) in order to estimate a more precise average effect.
However, the CREDO studies are well-known, and the 2013 national study covers more schools and a wider geography than any other study, so we include them as part of the overall body of evidence.
Table 1 communicates the strength of the methodology and confidence that the reader should place on the various studies: the darkest blue indicates lottery studies and meta-analyses, which generally provide the strongest evidence. Light blue indicates matched comparison studies, which generally provide somewhat less strong evidence. White indicates the CREDO “virtual peer” studies. (See “Measuring Effectiveness,” p. 10.)
With regard to outcomes, green in Table 1 indicates outcomes in which attending the charter schools in the study led to better outcomes (than attending a typical public school). Red indicates outcomes in which attending the charter schools in the study led to worse outcomes than attending the typical public school. Yellow indicates cases where attending the charter schools in the study made no difference relative to outcomes for those attending typical public schools. (Empty cells indicate that the outcome was not measured in the study in question.)
Taken together, the studies in Table 1 suggest the following takeaways:
“Average” charter school performance hides wide variation among individual charter schools
Just as traditional public schools range widely— from high-performing schools to “failing” ones—so, too, with charter schools. From a policy perspective, the wide variation in charter school performance indicates that policy makers, parents, and educators should look beyond averages to focus on individual charter schools—and what distinguishes effective ones from the rest. The second-to-last column of Table 1 indicates that, with the exception of the Mathematica study and the very geographically focused Boston studies, all the other studies found large variation—known technically as “heterogeneity”—in outcomes across charter schools. In less technical terms, this means that the studies found that charter schools differed from each other almost as much as they differed from noncharter schools when it comes to performance.
In general, the evidence indicates that charter schools generate gains in math and English for students… but with some exceptions
A quick scan of the test score columns in Table 1 (columns 5 and 6) illustrates how our understanding of charter schools’ effectiveness has evolved over the last decade. In general, the earlier studies, including the 2010 IES study covering charter schools in 15 states, showed little or no impact of charter schools on students’ math and English outcomes. In contrast, more recent studies (those toward the bottom of the chart) tend to show charter schools generating gains in student math and English outcomes relative to traditional public schools, with some exceptions.
Among the exceptions, CREDO’s 2014 study of Ohio’s charter schools found that, on average, charter school students achieved worse results than their “virtual peers” in traditional public schools, demonstrating the equivalent of 43 days less learning in math and 14 days less learning in reading. More recently, Dobbie & Fryer’s 2016 study of 45 Texas charter schools found that, on average, the schools had no impact on students’ math and reading scores.
While the Ohio and Texas findings may not be heartening to charter school proponents, they are consistent with takeaway number one, above: namely that great variability exists in charter school outcomes. When the Texas study disaggregated the results between five No Excuses charter schools and 40 “regular” charter schools, the five No Excuses charter schools generated gains for their students in reading and math (relative to typical public schools) and the 40 “regular” charter schools showed losses in reading and math (relative to typical public schools). The overall average in Texas (i.e., no impact of charter schools on student performance) hid the positive performance of the five No Excuses schools within the below-par performance of the other 40 charter schools. (“No Excuses schools” refers to schools that combine strong discipline, a commitment to math and reading achievement, and that devote substantially more time to instruction, whether through longer days or an extended school year.)
Recent, high-quality studies of charter schools show positive long-term impacts on students.
Methodology and strength of evidence of various studiesClick here to view the full table
Taken together, these studies suggest that urban charter schools are particularly effective in boosting students’ achievement.
With regard to Ohio, Ohio’s original charter school law has been criticized for having weak oversight provisions with regard to charter school authorizers. (Ohio strengthened its law in 2015, which postdates the study shown in Table 1.)48 Just as performance varies greatly among individual charter schools, some jurisdictions are likely to have higher quality charter schools (e.g., Boston) while other jurisdictions will have weaker charter schools (e.g., Ohio). CREDO’s 2014 report on Ohio stated, “…over 40 percent of Ohio charter schools are in urgent need of improvement.”
Recent studies suggest that charter schools hold promise for improving longer-term educational and employment outcomes for students from their late teens through mid-twenties
Starting in 2014, studies began being published that compared longer-term outcomes for charter school students with outcomes for comparable students who attended typical public schools. At this point the studies are few enough and the results are mixed enough, that it is still too early to reach firm conclusions. However, the pattern of evidence is tentatively promising.49 Examples include:
A 2014 study of schools in Florida and Chicago found that charter school attendance increased high school graduation rates, college enrollment rates, persistence in college (i.e., number of semesters completed) by nontrivial amounts, and—most intriguingly— average annual earnings by 12.7 percent for Florida charter school attendees when they were in their midtwenties ($20,780 versus $18,433 for students who attended noncharter public schools). Equally intriguing, the study found these positive long-term outcomes despite the fact that these same charter school students had not had better math and reading scores than their peers in traditional public schools when they were still in school. In short, the long-term impacts showed up despite the absence of earlier short-term impacts on educational achievement.
A 2017 study of schools in Chicago—where 22 percent of public high school students attend charter schools—found no impact on high school graduation rates (compared to traditional public schools) but strong positive impacts on the probability of enrolling in a four-year college and on college semesters completed.
Once again, the Texas study showed mixed results, with the five No Excuses charter schools leading to higher enrollment (relative to similar students at traditional public schools) in four-year colleges. However, the 40 “regular” charter schools led to decreased enrollment in four-year colleges and increased enrollment in two-year schools (relative to similar students at traditional public schools). Attending one of the 40 “regular” charter schools actually decreased earnings (relative to similar students at traditional public schools) when the individual reached his or her midtwenties. Attendance at the five No Excuses charter schools had no effect on earnings (relative to attendance at a traditional public school) during the students’ midtwenties.
Given the odd pattern of findings—positive earnings effects in Florida paired with no difference in earlier test scores; positive impacts 15 Policy Brief on test scores at the Texas No Excuses schools, but no impact on later earnings—it is tempting to speculate on what is driving these results. One possibility is that some charter schools are helping students develop “behavioral skills” that pay off in the labor market, and that these skills differ from the academic skills and knowledge that boost test scores in math and reading. The Mathematica report speculates that “small, mission-driven high schools of choice may be especially wellsuited to promoting” students’ long-term outcomes. Unfortunately, the current evidence remains too limited to know for sure. Hopefully future research will help illuminate these and other questions.
Time dedicated to educational instruction, comprehensive and strictly enforced codes of behavior, and a mission that prioritizes academic achievement are frequently cited attributes of successful charter schools.
Some hints of the types of charter schools that are most effective and the types of students that most benefit from them
The studies included in Table 1 were generally designed to answer whether charter schools benefited the “average” student attending them— not particular subgroups of students. Nevertheless, taken together, these studies begin to offer some hints of the types of charter schools that are most effective and the types of students who show the most benefit. They help move us beyond the firstorder question of “Are charter schools effective?” to the second-order question, “What types of charter schools are most effective, under what circumstances, and for which students?”
Taken together, these studies suggest that urban charter schools are particularly effective in boosting students’ achievement, relative to charter schools in rural or suburban areas. (See The Urban Charter School Performance Advantage for possible reasons for this result.)
Time dedicated to educational instruction, comprehensive and strictly enforced codes of behavior, and a mission that prioritizes academic achievement are frequently cited attributes of successful charter schools.50 Time dedicated to instruction usually means a longer school year or longer school days but can also be achieved through the inclusion of intensive tutoring.51 With regard to prioritizing academic achievement, many charter schools make meeting and exceeding math and reading proficiency standards a core part of their charter, and some go as far as to have parents and children sign contracts to that effect.
In terms of the types of students that most benefit from charter schools, multiple studies in Table 1 suggest that minority students (particularly African-American students), low-income students, or students with low educational achievement before entering the charter school show the greatest gains from attending one. Also reported, but less commonly, is that English-language learner (ELL) students, special education students, and boys benefit from charter school attendance. More research is needed before we can firmly conclude that these patterns hold up. A possible explanation underlying this pattern of results may be that charter schools are most effective for students who face the greatest challenges for academic achievement. The differences that charter schools can offer—whether in terms of instructional model, length of school day, size, or culture of the school—may help students for whom traditional public schools have not been effective.
Do charter schools adversely affect traditional public schools and the students who attend them?
As noted above, critics of charter schools argue that by diverting resources from traditional public schools, charter schools adversely affect those schools and the students who remain in them. In contrast, advocates of charter schools argue that charter schools help the performance of all public schools—charter and noncharter—by injecting positive competition into the system.
With the passage of time, and more experience with charter schools across multiple jurisdictions, we are now better able to begin resolving this debate.
Eleven studies of this question from 2004 to 2016—covering 11 different cities and states and one nationwide study—support advocates’ view that charter schools boost achievement for all public school students or at least have little or no negative impact on the achievement of students in the local noncharter schools. Six of the studies found that the presence of charter schools improved achievement for students at the local noncharter public schools. Four studies showed no effect of the presence of charter schools on the achievement of students in the local noncharter schools. Only one study—of an unnamed school district in the Southwest—found some negative effects of charter schools on the achievement of students in the local noncharter schools.52
The Urban Charter School Performance Advantage
One pattern that emerges from the research on charter schools is that charter school effectiveness varies with the location of the school—urban charter schools tend to be more effective than nonurban charter schools—and for different groups of students.* Low-income, minority students attending charter schools generally make greater academic progress than their white peers do when attending charter schools.
Given that charter schools’ “defining” features—principally, that they are schools of choice and enjoy more autonomy and greater freedom to innovate—are not dependent on location or students’ socioeconomic status, what accounts for this variation?
One possibility is that individual schools typically improve as they “mature.” Because charter schools were first established and remain concentrated in cities, urbanlocated charters might be of higher quality as they have had more time to improve their educational model. Another possible explanation for the apparent greater effectiveness of urban charter schools (compared to nonurban charter schools) is the student population that urban districts serve. Because students enrolled in traditional district schools in large cities tend to perform below average on standardized tests, urban charter schools could simply appear measurably more effective than nonurban charters because it is easier to help students with lower baseline scores quickly make substantial learning gains.
Urban and nonurban charters could also simply be structured differently based on parent demand. For instance, a suburban charter may respond to community demand for a nontraditional, non-testfocused environment, whereas urban charters are structured to meet demand for more structured, test-focused curricula. Although anecdotal, parental complaints over schools that “teach to the test” or stifle creativity are likely more common in districts where students are already meeting standardized test expectations—an area in which suburban-based schools, on average, generally excel.
Why students of different socioeconomic backgrounds might fare differently in charter schools is more complex to unpack, although many of the explanations for the variation between urban and nonurban charter school outcomes could help to explain variation by student background as well. For instance, if white, middle class students are more likely to attend high-quality traditional public schools, they will have comparably less to gain from attending an alternative school.* Center for Research on Education Outcomes at Stanford University, Urban Charter School Study: Report on 41 Regions, 2015.
Recommendations for Future Charter School Policy
As highlighted in this brief, charter schools across the country share the attribute of operating under a charter, but differ greatly in their organizational structure, culture, funding, types of students they serve, educational priorities, and most importantly, in their effectiveness. The early years of the charter school movement focused primarily on expanding the number of charter schools. The sector is growing and continues to solidify its position in American public education. Now, as student outcomes from this first set of charter schools indicate wide variation in the effectiveness of individual charter schools, it may be time for policymakers to shift their focus from the size of the sector to the quality of particular schools. Moving forward, policymakers, including legislatures, state education agencies, and charter school authorizers, should prioritize:
Improving quality control. The overall quality of the charter school sector, as measured by student outcomes, will become increasingly important, especially if the sector continues to expand. Just as with traditional public schools, individual charter schools vary widely in quality. Local and state school boards and governments are expected to intervene when traditional schools are failing. So, too, must charter school authorizers work to ensure that individual charter schools are providing a high-quality education. States, local governments, and authorizers should work in tandem to close under-performing or mismanaged charter schools and to develop best practices and tighter quality control when a charter is initially granted to a school.53
Providing high-quality charter options where population density supports growth. Highquality charter school options that support school choice and meet the varied educational needs of more children should be promoted where demand is sufficient to support additional schools. Regions with few school-aged children simply may not be able to support multiple schools without financially straining both traditional and charter schools.
Promoting charter schools as laboratories of innovation. One of the earliest arguments for charter schools was the idea that charter schools would serve as “innovation labs” to pilot and test new educational models. The resulting discoveries and insights would benefit all schools and students—not just those attending charter schools. This vision has not quite come to fruition. The wide variation in how charter schools approach education—and the results they achieve—offers an opportunity to learn much more about which practices work, under what conditions, and for which students. Those who care about student outcomes should embrace the idea that charter schools are a complement to rather than a competitor of traditional public schools. Currently traditional public schools educate 95 percent of children enrolled in American public schools. They will likely continue to educate the vast majority of children into the foreseeable future. It is in the nation’s best interest both to offer educational options and to learn as much as we can about effective educational practices wherever they can be studied.
- Most charter schools are run independently from the local school district. However, in some jurisdictions, such as the Los Angeles Unified School District (LAUSD), some charter schools are district-run. While this brief focuses on the conventional charter school model—schools that are owned and operated by nondistrict entities—it is important to recognize that the laws of any given state may permit a variety of public schooling models to develop that do not fit neatly into any predesignated school category. For information on how LAUSD differentiates between “affiliated” and independent charter schools, see “Public School Choice: School Models,” Los Angeles Unified School District.
- “Fast Facts: Charter Schools,” Institute of Education Sciences, National Center for Education Statistics (NCES).
- Many school districts allow parents to exercise school choice within the traditional public school sector by permitting interor intra-district transfers to other public schools. Despite many having the option to transfer, however, most students attend the neighborhood school assigned to them. “Fast Facts: Public School Choice Programs,” Institute of Education Sciences, National Center for Education Statistics (NCES).
- The application process is unique to each school; however, in some regions, charter schools have opted to accept a standard application to make the process easier for families applying for admission at multiple schools. For example, more than 80 charter schools in the greater Philadelphia region have opted-in to a standard application. See, “Philadelphia Charter School Standard Application,” Greater Philly Schools, July 2017.
- Kathy Christie, Maria Millard, Jennifer Thomsen and Micah Wixom, Trends in state charter school laws: Authorizers, caps, performance-based closures and virtual schools, Education Commission of the United States, June 2014.
- “Charter Schools: Background on Charter Schools,” National School Boards Association.
- For an example of the range of backfilling policies adopted by states and individual charter schools, see Backfilling in Charter Public Schools, National Alliance for Public Charter Schools, May 2016.
- Jack Buckley and Mark Schneider, Charter Schools: Hope or Hype? (New Jersey: Princeton University Press, 2009), p. 4.
- Center for Public Education (CPE), Charter schools: Finding out the facts: At a glance, March 2010.. For clarification on federal civil rights laws as they pertain to charter schools, see Catherine E. Lhamon, “Dear Colleague Letter: Charter Schools,” US Department of Education, Office for Civil Rights, May 14, 2014.
- “Charter Schools,” Minnesota Issues Resource Guides, Minnesota Legislative Reference Library, Minnesota State Legislature, July 2017.
- Ray Budde’s original proposal did not involve the creation of new schools; instead, he proposed district reform through the restructuring of existing public schools, with a focus on teacher-led education. See Ted Kolderie, “Ray Budde and the origins of the ‘Charter Concept’,” Education Evolving, Center for Policy Studies and Hamline University, June 2005.
- “Origins of Chartering Timeline,” Education Evolving, September 2010.
- “Origins of Chartering Timeline,” Education Evolving. While Shanker initially endorsed the idea of charter schools, he reportedly backed away from this endorsement as it became evident that the charter school sector was not developing in the way he envisioned, particularly with regards to whether charter schools must abide by district collective bargaining agreements. See Paul Peterson, “No, Al Shanker Did Not Invent the Charter School,” EducationNext, July 10, 2017.
- For more than 20 years, Albert Shanker wrote a paid weekly New York Time’s column, Where We Stand. For some of his early discussions on charter schools, see Albert Shanker, “Convention Plots a New Course: A Charter for Change,” New York Times [Paid Column], July 10, 1988..; Albert Shanker, “A Charter for Change, Con’t: Less TRUTH—Fewer Consequences,” New York Times [Paid Column], July 17, 1988.
- These figures reflect school and enrollment data for the 2014– 15 school year, the latest for which verified national data are available. See Table 216.20, 2016 Tables and Figures, Digest of Education Statistics, Institute of Education Sciences, National Center for Education Statistics (NCES), October 2016.
- 50-State Comparison: Charter School Policies, Education Commission of the United States, January 25, 2016.
- Table 216.30, 2016 Tables and Figures, NCES.
- “Student Enrollment,” DC Public Charter School Board (DC PCSB).
- Table 216.90, 2016 Tables and Figures, Digest of Education Statistics, Institute of Education Sciences, National Center for Education Statistics (NCES), October 2016.
- McKenna, Laura, “Why Don’t Suburbanites Want Charter Schools?” The Atlantic, October 1, 2015.
- District Performance Compared to the Nation, Institute of Education Sciences, National Center for Education Statistics (NCES), National Assessment of Educational Progress (NAEP). For more information on the data on urban district performance, which comes from the Trial Urban District Assessment (TUDA), see “About the Trail Urban District Assessment (TUDA),” National Assessment of Educational Progress (NAEP), March 3, 2016.
- Table 216.30, 2016 Tables and Figures, NCES.
- “Just the FAQs—Charter Schools,” Center for Education Reform.
- The Southern California Chapter of the American Civil Liberties Union (ACLU) has detailed examples of charter school practices that likely violate state or federal laws. Roxanne H. Alejandre, Angelica Jongco, Victor Leung, Unequal Access: How Some California Charter Schools Illegal Restrict Enrollment, American Civil Liberties Union of Southern California, August 1, 2016.
- Because so many of the nation’s authorizers are local school districts—according to the National Association of Charter School Authorizers, local school districts account for 90 percent of all authorizers—many charter school authorizers oversee only a couple charter schools, and sometimes, only one. See 2015 State of Charter Authorizing Report, National Association of Charter School Authorizers (NACSA).
- The National Association of Charter School Authorizers breaks down charter authorizers into six categories. For a list of active authorizers by state, see “Types of Authorizers,” National Association of Charter Schools Authorizers (NACSA). If a state is listed more than once, it has more than one type of active authorizer, although individual state laws will determine which schools can apply to which authorizers.
- Thomsen, Jennifer, “Charter Authorizers: What Are They and Why Do They Matter?,” Education Commission of the States (ECS), 2017.
- Sean Conclan, Alex Medler, and Suzanne Weiss, The State of Charter School Authorizing: 2009, National Association of Charter School Authorizers, pp. 45-8, May 6, 2010; Yilan Shen, Authorizing Charter Schools, National Conference of State Legislatures, May 2011.
- “Comprehensive Charter Public School Monitoring and Data Collection Processes,” National Alliance for Public Charter Schools.
- This figure represents charter application approval rates amongst authorizers overseeing 10 or more schools. These authorizers oversee more than 70 percent of all charter schools. “Inside Charter School Growth: A Look At Openings,” National Association of Charter School Authorizers (NACSA).
- While some caps can meaningfully restrict charter access, others are not restrictive from a practical standpoint. For instance, in California, there is a cap (2,500) on the total number of charter schools in the state, but that cap increases annually by 100; the current number of charters in the state (estimated at 1,253 for the 2016-17 school year) falls well below the total number currently allowed. “Charter Law Database: California,” National Alliance for Public Charter Schools.
- Principles and Standards for Quality Charter School Authorizing, National Association of Charter School Authorizers (NACSA), p. 14, 2015.
- A couple of states do not specify initial contract lengths in their charter laws. “Performance-Based Charter Contracts Required,” National Alliance for Public Charter Schools.
- The technical term for this is average daily attendance (ADA).
- Lack of public, transparent data makes comparing how much funding schools receive difficult; however, research does indicate charter schools on average receive less funding than traditional district schools, although how much less and whether it matters is subject to debate. Charter Schools In Perspective: A Guide to Research, Public Agenda and Spencer Foundation, Section 6, Finances. One frequently cited estimate is that charter schools receive approximately 30 percent less per student than traditional public schools. See Claudio Sanchez, “Just What IS A Charter School, Anyway?,” National Public Radio (nprEd), March 1, 2017.
- For an overview of how each state determines charter school funding, see “50-State Comparison: Charter Schools—How Is the Funding for a Charter School Determined?,” Education Commission of the States (ECS), January 2016.
- For example, in 2016, the Education Department awarded $245 million in new grants to support high-quality charter schools. See “U.S. Department of Education Awards $245 Million to Support High-Quality Public Charter Schools [Archived Information],” US Department of Education, September 28, 2016.
- Lisa Grover, Facilities Funding for Charter Public Schools, National Alliance for Public Charter Schools, September 2016.
- There is no comprehensive data on how much private aid charter schools receive. The amount of private funding a school attracts can vary greatly, although common sense tells us that large, well-known charter management organizations are more likely to receive private funds.
- Like traditional public schools, charter schools also have school boards, although charter schools typically have appointed (as opposed to elected) school boards.
- Having a for-profit substantially involved in the management of a charter school can be a divisive issue, even amongst charter advocates. See, Marian Wang, “When Charter Schools Are Nonprofit in Name Only,” ProPublica, December 9, 2014.
- James L. Woodworth, Margaret E. Raymond, Chunping Han, Yohannes Negassi, W. Payton Richardson, and Will Snow, Charter Management Organizations 2017, Center for Research on Education Outcomes (CREDO) at Stanford University, p. 18, 2017.
- 50-State Comparison: Charter Schools—Does the State Specify Who May Apply to Open a Charter School?, Education Commission of the States (ECS), January 2016.
- Woodworth, et al., Charter Management Organizations 2017, p. 35; Cynthia Hill, and David Welsch, Is there a Difference Between For-Profit Versus Not-For-Profit Charter Schools? [Working Paper], University of Wisconsin—Whitewater, Department of Economics, October 2007.
- As noted above, there are variations within these two types of schools. For instance, it is possible for several schools to have their entire operations overseen by a management organization that does not hold any of the individual school charters.
- One common distinction is to refer to nonprofit management organizations as Charter Management Organizations (CMOs) and for-profit organizations as Education Management Organizations (EMOs). This distinction is not universally Charting a Path Forward for Charter Schools applied, however, so it is important to remember that different groups may be referring to different management structures when referencing CMOs or EMOs.
- Calculations based on data from A Closer Look at the Charter School Movement, National Alliance for Public Charter Schools, p. 2. The Center for Research on Education Outcomes (CREDO) at Stanford University categories charter management organizations a bit differently, yet still find that about 32 percent of charter schools are a part of some type of larger managerial network. See Woodworth, et al., Charter Management Organizations 2017.
- Aaron Churchill, Jamie Davies O’Leary, and Chad L. Aldis, On the Right Track: Ohio’s Charter Reforms One Year into Implementation, The Thomas B. Fordham Institute, January 2017.
- While this pattern of results—little or no short-term impact on academic achievement but long-term impact on “life” outcomes, such as employment or earnings—may seem puzzling, it is in fact in keeping with research on other educational interventions. For example, the effect of early childhood programs on academic achievement and test scores typically fade away by approximately third grade. However, when these same individuals are tracked through adulthood, they often show positive impacts in terms of employment, reduced involvement with the criminal justice system, etc. One possible explanation is that these educational interventions are improving students’ “soft” or behavioral skills rather than the types of academic skills measured by standardized tests and other aspects of school achievement during the school-age years.
- Gleason, Philip, What’s the Secret Ingredient? Searching for Policies and Practices that Make Charter Schools Successful [Working Paper 47], Mathematica Policy Research, July 22, 2016; Matthew Di Carlo, The Evidence on Charter Schools and Test Scores, The Albert Shanker Institute, p. 4-5, December 2011.
- The effect of longer seat time is likely contingent on the extra being spent on core subject areas. Further, several studies find that when controlling for other factors, longer instruction time no longer has significantly positive impacts on student achievement. This may indicate that it is not, in fact, longer instruction time that is driving the increase in student achievement but one or more of the additional school policies and practices that are heavily correlated with time spent in school. However, it is also possible that time in school is so strongly correlated with these other factors that statistical analysis cannot separate out their individual effects. See Gleason, What’s the Secret, p. 11.
- Brian Gill, “The Effect of Charter Schools on Students in Traditional Public Schools: A Review of the Evidence,” Education Next, November 11, 2016.
- While having the ability to close charter schools is an important control mechanism, school closures can be highly disruptive for students and can have serious implications for learning if the students do not actually end up in better schools. To minimize these disruptions, policymakers should focus on ensuring only quality schools are improved in the first place.