School Choice and Charter Schools in Review: What Have We Learned?
School choice is front and center in the national debate on K-12 education in the United States. Secretary of Education Betsy DeVos has proposed to spend over $1 billion to expand school choice. DeVos has criticized the traditional public school system, referring to it as a monopoly, and has worked to expand school choice and charter schools in her home state of Michigan (Strauss 2017). Critics of school choice are wary of the privatization of schools and worry that school choice and charter programs harm traditional public schools and the children who attend them. The issue of school choice (and charter schools) remains highly politicized, with each side holding strongly to its views. Given several decades of experience with school choice and charter schools, can we assess the accuracy of some of the arguments in favor of and against school choice and charter schools? Further, how does the experiment with choice and charter schools inform our understanding of competition in non market environments?
Milton Friedman is often credited as beginning the national conversation around school choice with his 1955 essay, "The Role of Government in Education." Friedman posits that significant "neighborhood effects" warrant "each child to receive a minimum amount of education of a specified kind" (par. 6). These neighborhood effects, or positive externalities, exist because not only do the child and the parents receive the benefit of the child's education, but so do other members of society. As he states, "A stable and democratic society is impossible without widespread acceptance of some common set of values and without a minimum degree of literacy and knowledge on the part of most citizens" (par. 5). According to Friedman, government may therefore mandate some basic level of education. Such a requirement, however, would burden families financially. Therefore, Friedman proposes the use of government financial resources (i.e., tax revenues) to support primary and secondary education.
Government provision, or the use of tax revenues to provide education, however, does not necessarily imply government production, or what Friedman calls the "nationalization" of education. (1) Another option, which Friedman describes, is the voucher system.
Governments could require a minimum level of education which they could finance by giving parents vouchers redeemable for a specified maximum sum per child per year if spent on 'approved' educational services. Parents would then be free to spend this sum and any additional sum on purchasing educational services from an 'approved' institution of their own choice. The educational services could be rendered by private enterprises operated for profit, or by non-profit institutions of various kinds. The role of the government would be limited to assuring that the schools met certain minimum standards such as the inclusion of a minimum common content in their programs, much as it now inspects restaurants to assure that they maintain minimum sanitary standards (par. 11).
The status quo is that a child is assigned to a school based on where the child lives. In many cases, in order to change schools, parents must change their residence. Private schools are available, but at an additional cost. The key benefit of the voucher system is to give parents the opportunity to send their child to a wider range of schools, expanding the available choices. Friedman's argument in favor of vouchers and allowing private enterprise to produce education (with government ensuring that the schools meet certain minimum standards) ( 2002, p. 89) is centered on the value of freedom of choice.
In the decades following, school choice and charter programs were implemented across the country. School choice is designed to offer a voucher for students to attend private schools (including religious schools); most charter schools are public schools with higher levels of autonomy (see the distinctions across these school types in table 1). Researchers have attempted to evaluate the impact of school choice and charter schools by engaging with a range of research questions. For example, do students who attend school choice and charter schools perform better than students at traditional public schools? Or, do all students--across all school types-experience the benefits of competition? Do school choice and charter schools lead to a diversity of curricula and spur innovation in education? Do school choice and charter schools result in cream skimming, where the highest-achieving students move to one school and leave lower-achieving students behind? We address these key questions in section 3, where we summarize the findings in the literature on school choice and charter schools.
Several studies have summarized the arguments and findings on school choice, charter schools, and student achievement (measured in various ways) (Gill et al. 2001; Enlow and Ealy 2006; Rouse and Barrow 2008; Hess 2010). Hess provides an accessible history of the debate and highlights several major contributions of the last two decades. Enlow and Ealy's edited book addresses key questions within Friedman's argument and features a range of experts on school choice.
Our study contributes to the literature in three ways. One, we present and review responses to arguments in favor of and against school choice and charter schools and make a deliberate effort to reach across the ideological divide and consult a range of literature. We recognize the difficulties in this task and do not pretend to include all perspectives or offer an unbiased view. Two, we highlight some of the methodological and empirical issues that make some answers unclear. And three, although we address Friedman's 1955 essay, we focus our analysis on more recent contributions. For example, we analyze twenty quantitative studies published since 2002 on school choice and charter schools.
In the next section, we define the distinctions between school choice, charter schools, and what we refer to as "traditional public schools." We use Milwaukee, Wisconsin, as an example to further explain these different types of schools. Then, we present a series of propositions in favor and against school choice and charter schools. Finally, we offer a discussion of our findings and conclude.
II. School Choice, Charter Schools, and (Traditional) Public Schools
The first school choice program was established in 1990 in Milwaukee, Wisconsin. Today, fifteen states and Washington, DC, provide tax subsidies or publicly funded vouchers that allow students to attend private schools (EdChoice 2017). Many voucher programs allow students to attend religious schools, and the US Supreme Court established in Zelman v. Simmons-Harris (2002) that using vouchers for religious education does not violate the Establishment Clause of the First Amendment. Many school choice programs are means-based and targeted toward low-income students. School choice programs may practice selective admission. Arizona, Florida, Ohio, Pennsylvania, and Wisconsin have the greatest number of students that use vouchers to attend private schools, while Arizona has the highest percentage of students receiving a voucher, at approximately 6.3 percent of total K--12 students (Wolf 2012, p. 2).
Charter schools also arose during the early 1990s and are publicly funded schools that operate with some level of autonomy over curriculum--and, in some cases, over budgets and personnel (see table 1 for a comparison of charter schools in Milwaukee). As the National Alliance for Public Charter Schools states, "Charter schools are unique public schools that are allowed the freedom to be more innovative while being held accountable for advancing student achievement" (National Alliance for Public Charter Schools, n.d.). The National Alliance also states that charter schools are "open to all children; do not charge tuition; and do not have special entrance requirements." If a charter school has more applications than spots available, it must use a lottery system to allocate spots. Although charter schools cannot use (positive) student achievement to determine who they accept, there is likely a sorting effect, and it may be that more educated families apply (Ravitch 2016, p. 143).
Charter schools can be started by any person or organization. When applying to start a charter school, the applicant must write a proposal that outlines the curriculum and goals. An authorized agency reviews the charter school application and can either grant a charter (which typically extends for several years), or deny it. The school has a certain number of years to produce results that satisfy the agreement. Today, there are over 6,400 charter schools that serve over 2.5 million children, or approximately 5 percent of students in the United States (CREDO 2015).
As Ravitch (2016, p. xviii) states, charter schools are less controversial than vouchers because they do not involve church-state issues. Additionally, some charter schools (i.e., instrumentality charter schools) also hire teachers (who are unionized) from the district. Both school choice and charter schools expand choice, and sometimes the terms "private school choice" and "public school choice" are applied (respectively) to describe the two programs. A major point of contention against charter schools and school choice is that both programs take funding away from traditional public schools. (3)
Indeed, many people do not consider charter schools to be public schools because they have a definition of "public school" that goes beyond "publicly funded." Traditional public schools, as we refer to them here, are open to all students in the catchment area, follow state guidelines for curriculum and hiring teachers, and are governed by an elected school board. Proponents of traditional public schools tend to emphasize that they are local, neighborhood schools operating through democratically elected school board representatives.
A. School Choice and Charter Schools in Milwaukee, Wisconsin The first school choice program was established in Milwaukee in 1990. Today, as was the case in the 1990s, Milwaukee has a high poverty rate (approximately 29 percent) and is consistently ranked the most segregated city in the United States. With a population of just under 600,000 people, Milwaukee is racially diverse: approximately 44 percent of residents are white, 40 percent African American, and 16 percent Hispanic or Latino (2010 US Census). Racial disparities exist in economic status, with just 12 percent of white children living in poverty, compared to 33 percent of Hispanic or Latino children, and 49 percent of African American children (Wisconsin Council on Children and Families 2013).
Milwaukee has been at the forefront of school choice with the Milwaukee Parental Choice Program (MPCP). The radical idea was proposed by Polly Williams, a Wisconsin state legislator, and Howard Fuller, who was the Milwaukee Public School Superintendent. They believed that parents should have choices in where they send their children. As Williams stated in a 60 Minutes segment, the Parental Choice Program was needed urgently for children whose needs were not being addressed in public schools. Williams and Fuller likely did not think their proposal would be the catalyst for the urban education reform we have seen over the last quarter of a century.
The MPCP was established in 1990, and since 1995, the vouchers have been able to be used at religious schools. According to the Wisconsin Department of Public Instruction, 28,188 students were enrolled in the 120 schools that are a part of the MPCP during the 2016-17 school year, which represents roughly 25 percent of students in Milwaukee (MPS 2017). A 2006-07 survey of principals found that approximately 80 percent of the 120 MPCP schools were religious schools, and Catholic schools represented 30 percent of the MPCP religious schools (Kisida et al. 2008, p. 10). The MPCP is funded from general purpose revenue and a deduction in state aid that would otherwise be paid to Milwaukee Public Schools (MPS) (Wisconsin Department of Public Instruction 2016).
In 1993, Wisconsin enacted a law that enabled school boards to establish charter schools. (4) As indicated in table 1, there are three types of charter schools in Milwaukee. MPS authorizes both instrumentality and non instrumentality charter schools. Instrumentality refers to charter schools that have teachers hired by MPS (and are unionized); while non instrumentality schools hire their own teachers. (5) Independent, or 2r charter schools, are authorized by the University of Wisconsin-Milwaukee and City of Milwaukee and have a high level of autonomy. (6)
III. Propositions in Favor of and Against School Choice and Charter Schools
Since Friedman, numerous scholars have argued in favor of school choice and charter schools on the grounds that they expand choice and create competition. Chubb and Moe (1990) describe how the institutional framework for traditional public schools does not allow for autonomy (and instead, creates bureaucracy) and is not responsive to parents and students. By introducing school choice, the authors argue that schools will have greater incentive to move toward effective organization and improve the quality of education. Hoxby (2003a) has noted that improvements in education because of competition should be evident in all school types--the notion that competition lifts all boats.
In response, opponents of school choice and charter schools have argued that only certain types of parents and students will take advantage of choice and that those who do not will be made worse off (Witte 1996; Ladd 2002). Relatedly, opponents worry that school choice and charter schools will lead to increased racial segregation--a concern that Friedman also expressed (Friedman 1955, par. 16). And, others, including Witte (1996), have emphasized the importance of democratic control and the ability of community members to vote for school board members.
Below, we explore the literature further through the analysis of five key propositions around school choice and charter schools. (7) In addition, we searched through the literature to understand whether there is a consensus regarding the outcome of school choice and charter schools on student performance. Our sources come from Google Scholar searches, recommendations from education experts, and references from other articles and books. We also located the top twenty cited articles on the topic of school choice and charter schools (forty in total) according to Web of Science citation records. From our total of over sixty articles and books, we narrowed the list to twenty based on whether the article/book (1) addressed our question (i.e., do school choice or charter schools lead to improved student outcomes?), (2) provided a quantitative analysis, and (3) was published since 2002. We sorted the twenty articles into three categories: those that found school choice or charter programs did lead to improved outcomes, those that found they did not, and those that were inconclusive. Eight of the articles answer in the affirmative, six suggest no improvement or worse outcomes, and six are unclear. (8) We do not pretend that this sample is representative of all the literature on the topic; however, we do contend that there is a lack of consensus in the literature.
1. School choice will improve student outcomes by introducing competition. These improved outcomes will be experienced by both students in traditional public schools and students in choice or charter schools.
This argument runs counter to the (arguably) more common framing of identifying winners and losers of school choice and charter schools. Students remaining in traditional public schools are described as possible losers because they lose the benefit of having higher-achieving students in the classroom--assuming that they leave for choice or charter schools--and because their schools have fewer resources, as funding is diverted to choice or charter schools.
The basic economic argument is that when competition is introduced, all schools will have to improve outcomes in order to ensure that they maintain or attract students. Competition, however, is not a switch that is turned on or off. There are several different factors that influence the amount of competition (and therefore the possible level of improvement in outcomes), including (1) how many spots are available through choice or charter schools and how many students qualify for the program (e.g., school choice programs are often need based), (2) to what degree traditional public schools lose funding with school choice and charter enrollments, (3) what level of funding follows the student to choice and charter schools, and (4) other barriers, such as access to transportation to school choice and charter schools (Lieberman 2006). For example, when school choice was first introduced in Milwaukee in 1990, the program was limited to only 1 percent of city student enrollment, making competition small. In Arizona, 5.3 percent of the state's non private enrollment was in charter schools in 2000, and district-sponsored charter schools receive the same level of funding per pupil as traditional public schools do (Hoxby 2003b).
The empirical research on the impact of competition on student performance across school type is mixed. Hoxby (2003b) finds that public school achievement, measured by test scores, did increase when competition was introduced through vouchers (Milwaukee, WI) and charter schools (Michigan and Arizona). Greene and Marsh (2009) find positive, although modest, increases in student performance in Milwaukee across choice and traditional public schools (also see Clowes 2008). In contrast, Bettinger (2005) and Geller, Sjoquist, and Walker (2006) find that the impact of competition is not significant.
In summary: The studies are mixed. These inconsistent findings may be related to the level of competition or to methodologies that do not fully incorporate the level of competition into the studies.
2. School choice and charter schools will offer a diversity of curricula (which parents value) and stimulate innovation in education.
Diversity in curricula and innovation in education are at the center of school choice and charter schools. Many states explicitly say that innovation is a goal of charter schools. (9) By design, these schools have greater levels of autonomy and can pursue varied curricula, teaching practices, and administrative structures. In addition, economic theory suggests that competition will lead to innovation, as schools seek to differentiate themselves and improve quality in order to gain students. Again, the amount of competition will impact the amount of innovation that arises.
Innovation can take a variety of forms. Lubienski (2003) assembles studies that have sought to examine innovation following the introduction of school choice or charter schools. He finds that innovation takes place on various levels, from the administrative level to the classroom level (which includes pedagogy and curriculum). Further, there is no agreement on what constitutes innovation. Innovation is used to refer to "replications of familiar practices, adoption of practices used elsewhere, or development of new practices" (Lubienski 2003, p. 407).
Lubienski (2003) finds that many charter schools in California have increased parental involvement by having parents sign contracts. Some have distinctive pedagogies: for example, adopting project based instruction (Lubienski 2003, p. 410). Dobbie and Fryer (2011) study the Promise Academy, a charter school in New York City, that combines aspects of the "no excuses" philosophy with robust community programs, including karate and dance lessons, after school tutoring, and health programs. (10) In Milwaukee, competition has led to some visible diversified curricula in both traditional public schools and private choice schools. In 1999, MPS began implementing changes to give parents more options. Increased competition led to additional Montessori schools, the creation of a new technical high school, more before and after-school programs, full-day kindergarten, and the establishment of small schools within large high schools (Clowes 2008, pp. 373-74). Bruce Thomson was the superintendent of MPS at the time of these changes and acknowledged that the competitive pressure from choice encouraged MPS to add these programs (Clowes 2008, pp. 373-74). MPS has continued to give parents choices, adding two dual-language immersion schools, where students are taught both in English and Spanish. MPS is also adding a dual-language Montessori school.
Admittedly, some reviews of charter schools are less enthusiastic about innovations. In Colorado, some studies found that pedagogies were familiar and traditional. The Clayton Foundation stated, "Charters were not serving as 'laboratories of innovation' in Colorado" (Lubienski 2003, p. 410).
In summary: School choice and charter schools have brought diversity of curricula to cities and states. However, the scope of innovation and diversity varies.
3. School choice will lead to less mixing of students, as "parents of a kind flock together" (Friedman 1955, par. 16).
Friedman wrote his essay a year after the United States Supreme Court decided Brown v. Board of Education, the court decision that made it illegal to have separate schools for white and African American students. Friedman says, "Given greater freedom about where to send their children, parents of a kind would flock together and so prevent a healthy intermingling of children from decidedly different backgrounds" (par. 16). However, he also notes that this happens because of existing housing patterns: "Particular schools tend to be peopled by children with similar backgrounds thanks to the stratification of residential areas" (par. 16).
Starting in the 1990s, there has been resegregation of schools, reversing the trend from the 1950s to the late 1980s. The re-segregation occurred around the same time school choice and charter schools began. As Frankenberg, Lee, and Or field (2003, p. 6) note, there are high levels of residential segregation for African Americans and increasing levels for Latinos. According to the 2000-01 National Center for Education Statistics Common Core Data, whites are the most segregated group and attend schools that are 80 percent white, on average. Overall, schools are becoming more nonwhite, reflecting changing demographics with increases in Latino and Asian populations.
There are several hypotheses concerning how school choice or charter schools could contribute to increased racial segregation. Opponents of school choice worry that white families, who may be more educated, will take advantage of choice and send their children to private schools (because they perceive them to be higher quality). Or, parents of any race may decide where to send their children based on the school's racial composition. Alternatively, parents may have different preferences about what they want in a school, and if those preferences are correlated with race, this could lead to increased segregation. Finally, social capital, specifically related to the transfer of knowledge across social networks, is another way school choice could lead to increased segregation. If churches or community organizations inform their parishioners and members about the quality of various schools and those churches and organizations contain similar people, then there may be homogeneity in schools.
Empirical studies on the impact of school choice and charter schools on racial segregation are mixed, although more report increases in racial segregation. Bifulco and Ladd (2006, pp. 37-38) find that in North Carolina, an African American student will transfer from a traditional public school that is around 50 percent African American to a charter school that is more than 70 percent African American. White students that attend charter schools also have more racially homogenous peers; 24 percent of peers in a traditional public school are African American compared to 18 percent in charter schools (Bifulco and Ladd 2006, p. 38). Weiher and Tedin (2002, p. 79) report that in Texas, "Whites, African Americans, and Latinos transfer into charter schools where their groups comprise between 11 and 14 percentage points more of the student body than the traditional public school."
In contrast, Egalite and Mills (2014) examine the impact of the Louisiana Scholarship Program (LSP), a voucher program that allows low-income students in low-performing public schools to enroll in a private school. The authors find that "private schools with LSP enrollees are half as likely as public schools to be identified as racially homogeneous, which we define as having 90 percent of students belonging to the same race or ethnicity. Just 17 percent of LSP schools are racially homogeneous, compared to 34 percent of public schools that previously enrolled LSP students." Greene, Mills, and Buck (2010) review the impact of the MPCP on racial segregation and find that the program is neutral.
Still, there is debate about how to measure racial segregation. Greene and Mellow (2000) suggest that looking at school-level demographics is not enough; instead, studies should consider the racial composition of classrooms and interracial mixing in common areas (for example, lunchroom cafeterias). (11)
In trying to explain what drives increased racial segregation, Weiher and Tedin (2002) survey over one thousand parents of charter school students in Texas and ask how they determined which school to send their child to. The researchers offer six possible survey answers: test scores, discipline, common race/ethnicity, location, moral values, and safety. When they compare the ranking of "Common race/ethnicity" by different racial groups, it was ranked fourth and fifth, indicating that it is not that important. Different racial groups do exhibit different preferences, however. Among whites, test scores consistently ranked number one. African American parents ranked moral values as first, and Hispanic parents answered that discipline was their primary concern. Ball and Vincent (1998) conduct 138 interviews with parents and find that social capital (or, as the authors refer to it, the grapevine and "hot" knowledge) plays a significant role in where parents decide to send their children. To the extent that these social networks are racially/ethnically homogeneous, social capital could lead to increased segregation.
In summary: The majority of the literature suggests that school choice and charter schools lead to more racially segregated schools. Some studies report the opposite.
4. Students with parents who do not take an active interest in their child's education will be made worse off as high-achieving students leave the public school.
Proposition 1 in favor of school choice and charter schools posits that competition will "raise all boats," or that students in all types of schools will experience improvements in education. This is a Pareto improvement: all parties are made better off, and no one is made worse off. The proposition that "students with parents who do not take an active interest in their child's education will be made worse off" suggests that not everyone will benefit from competition through choice and/or charter schools. Critics of choice and charter schools worry that these schools will "skim the cream," or take the best students and leave the struggling students behind in the traditional public school (see, for example, Ladd 2002).
"Cream skimming" is troubling for a number of reasons, including the presence of peer effects in learning environments. Positive peer effects can take place when high-performing students encourage their classmates in the learning process. Negative peer effects occur then when students impede their classmates' learning because of low performance and possibly behavioral issues that disrupt classroom learning. (12) "Cream skimming" has the potential to remove high-performing students from the classroom, thereby limiting the opportunities for students in traditional public schools to benefit from the presence of high-achieving students.
Another important piece to consider when addressing this proposition is the student admission rules for choice and charter schools. School choice programs (with vouchers created to allow additional students to attend private schools) may practice selective admission. According to law, however, charter schools cannot practice "positive" selective admission, meaning that they cannot give preference to high-achieving students, but can give preference to low-achieving students. The rechartering process provides an opportunity to review student admission. If demand exceeds the number of spots available, a lottery determines admission.
Witte (1996) and Witte and Thorn (1996) have studied who benefits from the MPCP and other choice options. The authors are careful to point out that it depends on the rules in place, which vary by state. At the time Witte and Thorn conducted their study, they found that Chapter 220, (13) an integration program in operation since 1976 in Milwaukee, was pulling more high achievers from MPS compared to MPCP. Notably, MPCP had income requirements of 175 percent of the poverty level, while Chapter 220 admission was based on the racial composition of the assigned public school and could exclude students with attendance or behavioral problems.
Access to transportation may be another important factor that effectively excludes certain students. School choice programs do not typically provide transportation, and many charter schools also do not have to provide transportation. When transportation is not provided, families that have greater economic means and a reliable mode of transportation have an advantage (Zimmer et al. 2008, p. 2). Studies show that lack of transportation does prevent some parents from utilizing school choice and charter school options (see, for example, Teske, Fitzpatrick, and O'Brien 2009).
Recall that the empirical findings described in proposition 1 in favor of school choice and charter schools show that some studies find improvements across all types of schools (Hoxby 2003a; Greene and Marsh 2009; Clowes 2008), while other studies find no significant impacts from competition (Bettinger 2005; Geller, Sjoquist, and Walker 2006). Other studies have investigated whether cream skimming occurs, or whether students in traditional public schools are made worse off. Zimmer et al. (2008) examine data on students from seven different locations from 1997-98 to 2007-08 and find that charter schools do not systematically engage in cream skimming. Altonji, Huang, and Taber (2015) develop a model of the cream skimming effect and find that "the cream skimming effect of a voucher program on high school graduation rates [is] typically small in absolute value" (p. 320).
In summary: The rules in place around admission will have an important impact on whether "cream skimming" occurs. Most studies suggest that students in traditional public schools are not made worse off.
5. Private schools and charter schools do not have the same political legitimacy as traditional public schools. Further, private and charter schools may not contribute to "the common core of values deemed requisite for social stability." (Friedman 1955, par. 12)
Knight-Abowitz states, "For a school to be legitimate in a democratic society, its governance should be guided by democratic principles" (2013, p. 14). She goes on to explain that "citizens can become involved in this type of governance when schools meet two important requirements." First, education must prepare individuals for self-governance. "This principle requires a substantive education in citizenship, in which students learn the knowledge, skills, and values that prepare them for life in a democratic, pluralistic society in a globalized world." Second, "all citizens of a particular ward or district should be able to become involved with the governance and direction of schooling itself." This second point is emphasized by opponents of school choice and charter schools.
The most straightforward way that citizens are involved with the governance of schooling is by voting for school board members. School board members make decisions about curriculum, hiring, and the direction of the district. Opponents of school choice and charter schools argue that there is not broad democratic participation in private schools or charter schools. Private schools do not fall under the control of elected school boards, and charter schools have varying levels of autonomy. Further, there is a concern that charter schools are sometimes run by outsiders with no stake in the community.
Knight-Abowitz does recognize that many voters do not participate in local elections. She is also cognizant of arguments that "public schools are unnecessarily bureaucratic--and more responsive to the regulations of governance than to the concerns of local citizens, students, and educators themselves" (2013, p. 2). (14) For Knight-Abowitz and proponents of traditional public schools, the answer is to work toward reforms that address these issues without fundamentally changing the character of traditional public schools.
In 1988, two years before the MPCP was founded, Howard Fuller, advocate of school choice and charter schools in Milwaukee, stated, "We have a school system that is not working. It is failing to educate children, particularly poor Black children" (Fuller and Page 2014, p. 204). Fuller was fed up with the political process and bureaucracy. Fuller explained that if people could make the necessary changes to the existing system, "God bless you," but he was not willing to wait for those changes to take place (Fuller and Page 2014, p. 247). For Fuller, school choice represents a way for poor African American children to receive a better education. He also supports charter schools and has helped to develop more than thirty in Milwaukee. Although anyone can create a charter school, Fuller's vision is that local community members (in some cases, in cooperation with community outsiders) will come forward to accept the challenge.
Knight-Abowitz's first principle for political legitimacy, that education prepare students for citizenship, is, in fact, very similar to Friedman's argument on why a minimum level of education should be required for all. Recall Friedman's argument on the neighborhood effects of education: "A stable and democratic society is impossible without widespread acceptance of some common set of values and without a minimum degree of literacy and knowledge on the part of most citizens" (Friedman 1955, par. 5). Indeed, this defense of public education is hardly new. Others, including John Dewey, philosopher and education reformer, wrote extensively on the relationship between education, society, and democracy (see, for example, Dewey 1916).
This line of argument presents the questions of what such a curriculum for citizenship (including knowledge, skills, and values) looks like, and why might public, private, or charter schools be more or less effective at instilling these values or preparing students for citizenship?
A school's curriculum may promote these ideals by teaching the importance of individual rights and tolerance, the Constitution, and US history. It might also be the structure or pedagogy embedded in schooling that teaches the important lessons. It could be, for example, that certain classroom structures encourage collaboration and compromise, that the pedagogy promotes critical thinking, and that student governance (i.e., student council) inspires democratic participation.
In traditional public schools, it is easier to impose a common curriculum that teaches citizenship because there is government control of curriculum. Opponents of school choice and charter schools worry that, especially in the case of religious schools, the values that are taught may not prepare students to live in a diverse society. Some studies suggest otherwise. Wolf (2007), for example, reviews the literature on political tolerance and public and private schooling in the United States. He finds that ten out of eighteen studies show a positive effect of private education on political tolerance, and seven are not conclusive. Further, Coulson (2006) offers historical cases in which private religious schools of different sects coexisted within stable, diverse communities.
In summary: What is considered important for political legitimacy varies by person. All types of schools potentially offer ways to prepare students for citizenship (and again, it is unclear what preparation for citizenship looks like). IV. IV.
IV. Discussion and Conclusion
We have identified several arguments in the literature on school choice and charter schools and have pointed to places where there is emerging consensus and areas where there is no such consensus. Many propositions that we investigate have unclear results (or are unanswerable--for example, proposition 5). Notably, there does appear be greater diversity in curricula with the implementation of school choice and charter schools (proposition 2). There also appear to be more studies that suggest schools around the country are becoming more racially segregated and that school choice and charter schools contribute to the growing homogeneity in terms of race (proposition 3). These findings do not support the strong views about school choice or charter schools that voters and elected officials often hold. Instead, the empirical evidence suggests that the issue is more complicated. Below, we offer two explanations for this lack of consensus. First, we point to several empirical challenges that could explain some differences in findings. Second, we explore the diversity that exists within school choice and charter programs.
The unclear impact of school choice and charter schools may be due, in part, to empirical challenges in measurement, such as omitted variable bias, or to issues related to non observable variables, the appropriateness of various control variables, or challenges with longitudinal data. Since Coleman, Campbell, and Hobson's (1966) seminal study, researchers have been careful to include socioeconomic status variables as controls when considering the impact of school choice or charter programs on student achievement. Still, there could be non observable variables present among those families and students that pursue school choice options and those that remain in traditional public schools. Random assignment to a private school or charter school by lottery could eliminate this possibility by controlling for other variables. (A lottery system is often used when demand exceeds supply.) Researchers could then compare the lottery winners and the lottery losers and be reasonably sure that the differences in outcomes were a result of the treatment (i.e., type of school) and not an unobservable variable. (15)
In addition to the difficulties in identifying relevant variables that may impact student outcomes, there are also proxies used to represent, for example, household income, that are imprecise. A report by Peterson and Llaudet (2007) argues that National Assessment of Educational Progress (NAEP) data, published by the National Center for Education Statistics, suffers from a classification bias because "it infers a student's background from his or her participation in federal programs intended to serve disadvantaged students" without recognizing that "public and private school officials have quite different obligations and incentives to classify students as participants in these federal programs" (p. 76). Because of this classification bias, the NAEP data understate the level of need in the private sector and overstate need in the public sector. (16)
To understand the impact of a particular school on student outcomes, longitudinal data are especially helpful. Some studies compare test scores of cohorts in private or charter schools with test scores of cohorts in traditional public schools and compare the change in outcomes. This method presents two problems.
One is that students move in and out of a school, and therefore, unless student-level data are used, researchers end up tracking different groups of students. This is the same problem that Thomas Sowell (2011) has pointed to with looking at snapshots of income levels; the data do not accurately reflect the well-being of households or individuals over time because the composition of the group changes. Other studies do use student-level longitudinal data and use a matching strategy or student fixed effects. The matching strategy considers students from private or charter schools and finds a match, a student with similar characteristics within the traditional public school system. One drawback to this approach is that non observable variables are still present (related to the sorting to private/charter schools from traditional public schools).
A further explanation for the lack of consensus in the literature is that there is great diversity within school type. Table 1 presents (still at a high level) three different types of charter schools in Milwaukee and some of the institutional differences (e.g., teacher characteristics, the level of autonomy, and per-pupil funding). Unfortunately, many studies do not factor in differences within school type, and instead either use a dummy variable to indicate school type, or include one or two additional characteristics. By not investigating these other differences (for example, curriculum, how material is taught, etc.), we cannot really understand what is driving student outcomes.
Competition (outside of perfect competition) leads to innovation and product differentiation. To understand student outcomes, researchers must consider a host of characteristics beyond private, charter, or traditional public school. Some of these characteristics include teacher experience, teacher quality, curriculum (the academic content as well as the pedagogy and disciplinary policies), the age of the institution (e.g., a charter school that has been in operation for one year versus fifteen years), the resources available, whether transportation is available, and the other services that the school provides (e.g., after-school tutoring, daycare, etc.). Understanding how particular practices improve student outcomes is essential and ought to be the focus of future studies. Debates about school choice and charter schools are empty without a consideration of the diversity that exists within school type.
Abdulkadiroglu, Atila D., Joshua Angrist, S. Dynarski, P. J. Kane, and Parag A. Pathak. 2011. "Accountability and Flexibility in Public Schools: Evidence from Boston's Charters and Pilots." Quarterly Journal of Economics, 126(2): 699-748.
Altonji, Joseph, Ching-I Huang, and Christopher Taber. 2015. "Estimating the Cream Skimming Effect of School Choice." Journal of Political Economy, 123(5): 266-324.
Angrist, Joshua D., Parag A. Pathak, and Christopher R. Walters. 2013. "Explaining Charter School Effectiveness." American Economic Journal: Applied Economics, 5(4): 1-27.
Ball, Stephen J., and Carol Vincent. 1998. "'I Heard It on the Grapevine': 'Hot' Knowledge and School Choice." British Journal of Sociology of Education, 19(3): 377-400.
Bettinger, Eric P. 2005. "The Effect of Charter Schools on Charter Students and Public Schools." Economics of Education Review, 24(2): 133-47.
Bifulco, Robert, and Helen F. Ladd. 2006. "The Impacts of Charter Schools on Student Achievement: Evidence from North Carolina." Education Finance and Policy, 1(1): 50-90.
Booker, Kevin, Scott M. Gilpatric, Timothy Gronberg, and Dennis Jansen. 2007. "The Impact of Charter School Attendance on Student Performance." Journal of Public Economics, 91(5-6): 849-76.
Carter, Samuel C. 2000. "No Excuses: Lessons from 21 High-Performing, High Poverty Schools." Washington, DC: Heritage Foundation.
Center for Research on Education Outcomes (CREDO). 2015. "Urban Charter School Study Report on 41 Regions," Stanford, CA.
Chubb, John, and Terry Moe. 1990. Politics, Markets, and America's Schools. Washington, DC: Brookings Institution.
Clowes, G. A. 2008. "With the Right Design, Vouchers Can Reform Public Schools: Lessons from the Milwaukee Parental Choice Program." Journal of School Choice, 2(4): 367-90.
Coleman, James, Earnest Campbell, and Carol Hobson. 1966. Equality of Educational Opportunity. US Government Printing Office.
Coulson, Andrew. 2006. "A Critique of Pure Friedman: An Empirical Reassessment of 'The Role of Government in Education.'" In Liberty and Learning: Milton Friedman's Voucher Idea at Fifty, eds. Robert Enlow and Lenore Ealy, 103-23. Washington, DC: Cato Institute.
Cowen, Joshua M., David J. Fleming, John F. Witte, Patrick J. Wolf, and Brian Kisida. 2012. "Report #30, Student Attainment and the Milwaukee Parental Choice Program: Final Follow-Up Analysis." School Choice Demonstration Project.
Cullen, Julie B., Brian A. Jacob, and Steven D. Levitt. 2005. "The Impact of School Choice on Student Outcomes: An Analysis of the Chicago Public Schools." Journal of Public Economics, 89(5-6): 729-60.
Cullen, Julie B., Brian A. Jacob, and Steven D. Levitt. 2006. "The Effect of School Choice on Participants: Evidence from Randomized Lotteries." Econometrica, 74(5): 1191-1230.
Dewey, John. 1916. Democracy and Education. Mineola, NY: Dover Publications.
Dobbie, Will, and Roland Fryer. 2011. "Are High-Quality Schools Enough to Increase Achievement among the Poor? Evidence from the Harlem Children's Zone." American Economic Journal: Applied Economics, 3(3): 158-87.
EdChoice. 2017. "Fast Facts on School Statistics." EdChoice.org.
Egalite, Anna, and Jonathan Mills. 2014. "The Louisiana Scholarship Program." EducationNext, 14(1): 66-69.
Enlow, Robert, and Lenore Ealy, eds. 2006. Liberty and Learning: Milton Friedman's Voucher Idea at Fifty. Washington, DC: Cato Institute.
Epple, Dennis, and Richard E. Romano. 1998. "Competition between Private and Public Schools, Vouchers, and Peer-Group Effects." American Economic Review, 88(1): 33-62.
Frankenberg, Erica, Chungmei Lee, and Gary Or field. 2003. "A Multiracial Society with Segregated Schools: Are We Losing the Dream?" The Civil Rights Project. Cambridge, MA: Harvard University.
Friedman, Milton. 1955. "The Role of Government in Education." In Economics and the Public Interest, ed. Robert Solo, 123-44. New Brunswick, NJ: Rutgers University Press.
Friedman, Milton. (1962) 2002. Capitalism and Freedom. Chicago: University of Chicago Press.
Fuller, Howard, and Lisa F. Page. 2014. No Struggle, No Progress: A Warrior's Life from Black Power to Education Reform. Milwaukee: University of Marquette Press.
Gaviria, Alejandro, and Steven Raphael. 2001. "School-Based Peer Effects and Juvenile Behavior." Review of Economics and Statistics, 83(2): 257-68.
Geller, Christopher R., David L. Sjoquist, and Mary B. Walker. 2006. "The Effect of Private School Competition on Public School Performance in Georgia." Public Finance Review, 34(1): 4-32.
Gill, Brian, P. Mike Timpane, Karen E. Ross, Dominic J. Brewer, and Kevin Booker. 2001. Rhetoric versus Reality: What We Know and What We Need to Know about Vouchers and Charter Schools. Santa Monica, CA: RAND.
Gleason, Phillip, Melissa Clark, Christina C. Tuttle, and Emily Dwoyer. 2010. "The Evaluation of Charter School Impacts." National Center for Education Evaluation and Regional Assistance. Institute of Education Sciences. US Department of Education.
Greene, Jay P., and Ryan H. Marsh. 2009. "Report #11: The Effect of Milwaukee's Parental Choice Program on Student Achievement in Milwaukee Public Schools." The School Choice Demonstration Project.
Greene, Jay P., and N. Mellow. 2000. "Integration Where It Counts: A Study of Racial Integration in Public and Private School Lunchrooms." Texas Education Review, 1(1): 15-26.
Greene, Jay P., Jonathan Mills, and Stuart Buck. 2010. "Report #20: The Milwaukee Parental Choice Program's Effect on School Integration." The School Choice Demonstration Project.
Hanushek, Eric A., John F. Kain, Steven G. Rivkin, and Gregory F. Branch. 2007. "Charter School Quality and Parental Decision Making with School Choice." Journal of Public Economics, 91(5-6): 823-48.
Hess, Frederick. 2010. "Does School Choice 'Work'?" National Affairs, 33.
Hoxby, Caroline M. 2003a. "How School Choice Affects the Achievement of Public School Students." In The Economics of School Choice, ed. C. M. Hoxby, 141-77. Chicago: University of Chicago Press.
Hoxby, Caroline M. 2003b. "School Choice and School Competition: Evidence from the United States." Swedish Economic Policy Review, 10(2): 11-67.
Hoxby, Caroline M. 2004. "A Straightforward Comparison of Charter Schools and Regular Public Schools in the United States." Harvard University and National Bureau of Economic Research.
Kisida, Brian, Laura Jensen, James Rahn, and Patrick Wolf. 2008. "Report #3: The Milwaukee Parental Choice Program: Baseline Descriptive Report on Participating Schools." The School Choice Demonstration Project.
Knight-Abowitz, Kathleen. 2013. Publics for Public Schools: Legitimacy, Democracy and Leadership. Boulder: Paradigm.
Ladd, Helen. 2002. "Comment on Caroline Hoxby, 'School Choice and School Competition: Evidence from the United States.'" Swedish Economic Policy Review, 10: 67-76.
Lieberman, Myron. 2006. "Free Market Strategy and Tactics in K-12 Education." In Liberty and Learning Milton Friedman's Voucher Idea at Fifty, eds. Robert Enlow and Lenore Ealy, 81-101. Washington, DC: Cato Institute.
Lubienski, Christopher. 2003. "Innovation in Education Markets: Theory and Evidence on the Impact of Competition and Choice in Charter Schools." American Educational Research Journal, 40(2): 395-443.
Lubienski, Christopher, and Sarah Lubienski. 2014. The Public School Advantage: Why Public Schools Outperform Private Schools. Chicago: University of Chicago Press.
Merry, Michael S., and William S. New. 2016. "Is the Liberal Defense of Public Schools a Fantasy?" Critical Studies in Education, 58(3): 373-89.
Milwaukee Public Schools (MPS). 2017. "District Fact Sheet."
Minnesota Statutes, Education Code: Prekindergarten-Grade 12[section]124D.10, Sub. 1. (1991).
National Alliance for Public Charter Schools. n.d. "About Charter Schools." PublicCharters.org.
Ostrom, Vincent, Charles M. Tiebout, and Robert Warren. 1961. "The Organization of Government in Metropolitan Areas: A Theoretical Inquiry." American Political Science Review, 55(4): 831-42.
Peterson, Paul E., and Elena Llaudet. 2007. "The NCES Private-Public School Study: Findings Are Other Than They Seem." Education Next, 7(1): 75-79.
Ravitch, Diane. 2016. The Death and Life of the Great American School System: How Testing and Choice Are Undermining Education. New York: Basic Books.
Rouse, Cecilia, and Lisa Barrow. 2008. "School Vouchers and Student Achievement: Recent Evidence, Remaining Questions." Annual Review of Economics, 1(1): 17-42.
Sass, Tim R. 2006. "Charter Schools and Student Achievement in Florida." Education Finance and Policy, 1(1): 91-122.
Sowell, Thomas. 2011. Economic Facts and Fallacies. New York: Basic Books.
Strauss, Valerie. 2017. "What 'School Choice' Means in the Era of Trump and DeVos." Washington Post, May 22.
Teske, Paul, Jodi Fitzpatrick, and Tracey O'Brien. 2009. "Drivers of Choice: Parents, Transportation, and School Choice." Center on Reinventing Public Education. University of Washington, Bothell.
US Census Bureau. 2010. "Race and Hispanic or Latino Origin."
Weiher, Gregory R., and Kent L. Tedin. 2002. "Does Choice Lead to Racially Distinctive Schools? Charter Schools and Household Preferences." Journal of Policy Analysis and Management, 21(1): 79-92.
Wisconsin Council on Children and Families. 2013. WisKids Count Policy Brief, Fall.
Wisconsin Department of Public Instruction, 2016. "Wisconsin, Racine, and Milwaukee Voucher Enrollments Announced." News release.
Wisconsin Institute for Law and Liberty (WILL). 2015. "'Apples-to-Oranges?' When Comparisons Are Made between MPCP and MPS." WILL, 2(2): 1-7.
Witte, John. 1996. "Who Benefits from the Milwaukee Choice Program?" In Who Chooses? Who Loses? Culture, Institutions, and the Unequal Effects of School Choice, eds. Bruce Fuller and Richard Elmore, 118-37. New York: Teacher's College Press.
Witte, John, and Christopher Thorn. 1996. "Who Chooses? Voucher and Inter district Choice Programs in Milwaukee." American Journal of Education, 104(3): 186-217.
Wohlstetter, Priscilla, Michelle B. Nayfack, and Eugenia Mora-Flores. 2008. "Charter Schools and 'Customer' Satisfaction: Lessons from Field Testing a Parent Survey." Journal of School Choice, 2(1): 66-84.
Wolf, Patrick J. 2007. "Civics Exam." Education Next, 7(3): 67-72.
Wolf, Patrick J. 2012. "Report #36: The Comprehensive Longitudinal Evaluation of the Milwaukee Parental Choice Program: Summary of Final Reports." The School Choice Demonstration Project.
Zelman v. Simmons-Harris. 536 US 639 (2002).
Zimmer, Ron, Brian Gill, Kevin Booker, Stephane Lavertu, and John F. Witte. 2008. "Do Charter Schools 'Cream Skim' Students and Increase Racial-Ethnic Segregation?" Prepared for School Choice and School Improvement: Research in State, District and Community Contexts. Vanderbilt University.
Zimmer, Ron, Brian Gill, Kevin Booker, Stephane Lavertu, and John F. Witte. 2012. "Examining Charter School Achievement in Seven States." Economics of Education Review, 31(2): 213-24.
* The authors thank Jim and Marge Sanger for their support of the Sanger Summer Research Program, which made our study possible. We also thank Charles Westerberg, Evelyn Roman, Bill New, Jarett Fields, and Fabiola Ramirez for their helpful discussions on the issues related to school choice and charter schools. All remaining errors are our own.
(1) This distinction between provision and production is further explored by Ostrom et al. 1961, p. 834.
(2) For basic information about choice programs around the country, see National Conference of State Legislatures, "School Voucher Laws: State by State Comparison," NCSL.org.
(3) To differentiate between charter schools and non charter public schools, we use the term "traditional public schools" to describe the latter.
(4) According to the Office of Charter Schools (University of Wisconsin-Milwaukee School of Education), "The role of charter schools is to promote innovation, develop new models of education, and create working environments that foster improved educational opportunities for children. Charter schools offer a new governance structure for public schools that trades autonomy for accountability and holds high academic and organizational expectations for the school."
(5) The politics of unions have come under attack over the last ten years. In 2011, Wisconsin passed Act 10, effectively ending collective bargaining for teachers, the major benefit of unions. Notably, teachers' unions were plaintiffs in both the Wisconsin Supreme Court and the United States Supreme Court ruling that allowed vouchers to go to religious schools.
(6) Autonomy is defined as decision-making authority over curriculum, budget, and personnel (including the ability to hire teachers not part of a union).
(7) A referee has pointed out that a further proposition to explore is related to customer satisfaction and whether parents of children at school choice or charter programs express higher levels of satisfaction. Indeed, this is a question that has been considered in the literature. Interested readers may consult Wohlstetter et al. 2008, for example.
(8) All twenty studies are referenced at the end of the paper. Those that answer in the affirmative are Abdulkadiroglu et al. 2011; Booker et al. 2007; Cowen et al. 2012; Dobbie and Fryer 2011; Hoxby 2003a; Hoxby 2004; Sass 2006; and Wisconsin Institute for Law and Liberty (WILL) 2015. Those that report no improvement or worse outcomes are Bettinger 2005; Bifulco and Ladd 2006; Cullen et al. 2005; Hanushek et al. 2007; and Lubienski and Lubienski 2014 (two studies are used from this source). Those that have unclear outcomes are Angrist et al. 2013; Cullen et al. 2006; Geller et al. 2006; Gleason et al. 2010; and Wolf 2012.
(9) Lubienski (2003, p. 399) references the state of Minnesota, which established charter schools as vehicles to "(1) improve pupil learning; (2) increase learning opportunities for pupils; (3) encourage the use of different and innovative teaching methods" (Minnesota, 1991, 124D.10, Sub. 1).
(10) Dobbie and Fryer (2011) reference Carter's (2000) definition of "no excuses" schools as schools that allow the principal administrative freedom, set measurable goals that are regularly assessed, emphasize parent participation, and encourage a culture of universal achievement with "no excuses."
(11) Tracking systems in schools can result in homogeneity in the classroom even when the school's racial makeup is less homogeneous.
(12) For empirical studies on peer effects, see Epple and Romano (1998); Gaviria and Raphael (2001); and Hanushek et al. (2007).
(13) Chapter 220 is currently (2017) being phased out in Milwaukee.
(14) Also see Merry and New (2016).
(15) One challenge to this approach is that the lottery system is only used when there is excess demand for a school, which suggests that the private and charter schools that use lotteries may be of higher quality than other private and charter schools (Zimmer et al. 2012, p. 216). For more information on this strategy, also see Hoxby 2003a, pp. 146-52.
(16) In this literature review, we do include at least two studies that use this data set (see Hoxby 2004; Lubienski and Lubienski 2014).
Table 1. School types in Milwaukee, Wisconsin Milwaukee Choice Charter: Public Schools Instrumentality Public / Public Private Public Private Eligibility All Family income No income students up to 300% of requirements; poverty level if excess Enrollment demand, schools cap: none for must use MPCP; no lottery * selective admission; lottery Authorizer MPS State MPS superintendent Employees MPS School MPS Autonomy None High Limited Per-Pupil $10,122 $7,323 for Varies ** Funding, grades K-8; 2016-17 $7,969 for grades 9-12 Number of 160 121 5 Schools, 2016-17 Number of 66,381 28,188 1,635 Students, 2016-17 Charter: Charter: Noninstrumentality Independent (2r) Public / Public Public Private Eligibility No income No income requirements; if requirements; excess demand, if excess schools must use demand, lottery * schools must use lottery * Authorizer MPS Common Council/UWM Employees School School Autonomy High High Per-Pupil Varies ** $8,188 Funding, 2016-17 Number of 17 21 Schools, 2016-17 Number of 8,840 7,401 Students, 2016-17 Sources. See WILL 2015; on i school choice, see Wisconsin--Milwaukee Parental Choice Program, EdChoice.org, and Wisconsin Department of Public Instruction, "Private School Choice Programs: Frequently Asked Questions, 2016-17 School Year." Other data obtained from the Department of Public Instruction (DPI). Notes. MPCP is Milwaukee Parental Choice Program; MPS is Milwaukee Public Schools; UWM is University of Wisconsin-Milwaukee. * Under Act 55, changes to charter school law, (1) schools would be required to give preference to pupils enrolled in the charter school in the prior year and their siblings; (2) schools may give preference to children of the school's founders, governing board members, and full-time employees, but must limit the number of such children to no more than 10 percent of the charter school's total enrollment. ** "In schools chartered by a school district, the contract or charter controls the amount of funding the charter school receives each year. In some cases, the district's per-pupil expenditure follows the student as he or she moves from a regular public school to a charter school. In other cases, the charter school functions with less money." DPI Yearbook, 2016-17.