Why some places have more students in charter schools and others have fewer
By most measures, the charter school reform movement has been remarkably successful. Since the first law authorizing charter schools was passed in Minnesota in 1991, 39 other states, the District of Columbia, and Puerto Rico have all adopted legislation supporting public charters. Today, more than 1.2 million U.S. school children attend more than 4,000 public charter schools.
But the success of the charter school movement has been as uneven as it has been widespread (see Figure 1). There are remarkable differences in the number of charter schools and enrollment between states, and even between school districts within the same state. Take Arizona and Minnesota. The two states were early leaders in the charter school movement, both passing legislation highly favorable to the establishment and support of charter schools. Yet, in the 2005–06 school year, more than 10 percent of Arizona’s enrollment was in charter schools, while only 3 percent of Minnesota students attended a charter school.
The patchwork pattern of success for the charter school movement in the United States raised two big questions in our minds. What factors led some states to grant charter schools a great deal of latitude and provide solid financial support, while others adopted less permissive legislation? And, why, even among states with similar enabling legislation, do charter schools flourish in some places but not in others?
Several well-regarded researchers have tried to explain the differences in charter school legislation. We decided to build on their work, in an effort to produce a more complete account of the politics of the charter school movement. Like those conducting the previous studies, we considered the role of state demographics and party politics. We also used new data to see whether the academic performance of students in traditional public schools and the influence of teachers unions affect the strength of charter school legislation in a state. But this was only one part of our larger project. Marshalling demographic, financial, political, and school performance data from 1990 to 2004, we took the novel step of assessing patterns in the presence of charter schools and in their enrollments at both the state and local levels.
The first thing we needed to do was identify U.S. charter schools and their locations and determine their enrollments. The most recent comprehensive catalog of charter schools and school enrollments is the National Center for Education Statistics (NCES) Common Core of Data. We combined the NCES file with the Center for Education Reform (CER) directory of charter schools to create a master database of 3,066 schools for the 2003–04 school year. Next, we calculated the total number of charter schools and the total enrollment in charters and traditional public schools in each school district. In all, at least 1,000 of the 14,000 districts in the U.S. contained at least one charter school.
With this information in hand, we set out to answer three questions about the charter school movement at the state level. One, why did some states pass charter laws earlier than others? To answer this question, we first studied how D.C. and the 37 states that passed charter laws before 1999 differed from the remaining states that had not adopted charter laws by 1999. For the 40 states that passed a charter law by the 2003–04 school year, we also investigated how earlier and later adopters, grouped by year of the law’s enactment, differ from one another. We wanted to know, for instance, how Minnesota, the state that passed the nation’s first charter school law in 1991, is different from Maryland, which passed the most recent enabling legislation in 2003.
Two, we wondered why some states enacted laws highly favorable to charter schools while others passed more-restrictive statutes. In this part of our study, we compared states based on the rating of their laws by CER, which is an advocacy organization for charter schools. CER rates the “strength,” or permissiveness, of the laws’ provisions. CER judges each law against 10 criteria, each scored on a 1–5 scale, with a total possible score of 50 for laws most favorable to charter schools. These criteria include whether the state grants charter schools an exemption from collective bargaining, the number of chartering authorities beyond local school boards, the number of new charter schools permitted, and whether charters are granted waivers from certain state and local laws. Arizona, Michigan, and Minnesota have enacted relatively “strong” legislation, that is, legislation that provides considerable latitude to charter schools. Other states, such as Kansas, Tennessee, and Virginia, have adopted charter legislation with much more restrictive provisions. In our study, we assigned states without charter school laws a CER law strength score of 0. Although an advocacy organization may have an incentive to understate the strength of these laws, it is unlikely that any overall downward bias would create problems for our assessment of states relative to one another (and, in fact, these scores have been used in a number of other scholarly studies).
Finally, we asked why some states have a greater percentage of public school students enrolled in charter schools than other states. In all our analyses of state laws and statewide enrollment, we wanted to know how characteristics of the state affect the level of support for charters. We were primarily interested in assessing the role of demographics, student achievement, and the extent of school choice currently available in shaping how states participate in the charter school movement. Because the presence of charter schools in an area might affect both student achievement and the decisions of families to move to a district, we measured state demographics and student achievement during the 1989–90 school year, several years before the first charter laws took effect. When studying the pattern of charter school enrollment across the country, we took into account how each of three factors contributes to or retards charter school growth: per pupil expenditures (also measured during the 1989–90 school year), length of time a charter law was on the books, and degree of permissiveness of each state’s charter school law, as measured by the CER index.
Demographics and Politics
Before we proceed, let’s consider why we would expect demographics to shape charter school politics and participation. It is possible that different racial, ethnic, and economic groups demand different curricular approaches: for example, bilingual education, arts, vocational instruction, or programs for gifted or at-risk students. In areas with greater racial and ethnic diversity and economic inequality, families and policymakers may support charter schools as a means of satisfying diverse educational preferences. Indeed, many charter school founders explicitly state that satisfying the educational needs of a target student group is central to their mission. Or it may be that demographics matter because changes in the demographic composition of local school districts increase the desire among families to sort students into similar peer groups. Other researchers have found that white students in charter schools transferred from schools that, on average, had a higher proportion of nonwhite students than their new charter school. In any case, we expected states and communities with more heterogeneous populations to be more supportive of the charter school movement.
A word about our measure of student achievement. It was not clear to us whether parents are most concerned with the absolute level of student performance or with performance relative to what they expect given resources in the school, household, and community. For example, expectations about high-school dropout rates could vary with K–12 expenditures and local poverty. Whether parents, and policymakers, are ultimately “satisfied” with public education may hinge on how well schools are performing relative to local expectations. We conducted our analysis alternately using absolute student achievement, measured with statewide mean SAT scores for the 1989–90 school year and the mean high-school dropout rate calculated from 1990 census data, and with a second measure that represents the deviation of actual achievement from expected student performance.
The two achievement measures, SAT scores and dropout rates, have the benefit of reflecting student achievement at both ends of the ability distribution. A disadvantage, however, is that SAT scores and dropout rates are much more closely aligned to secondary-school performance than to elementary-school performance. And only about 25 percent of charter school students were enrolled in a secondary grade during the 2003–04 school year. We did, however, have a reasonable amount of confidence in the use of these achievement measures because mean SAT scores and high-school graduation rates are very public indicators of educational outcomes, and parents and policymakers are likely to consider these measures when they take positions on charter schools.
To estimate expected student performance, we used a statistical procedure that predicts student achievement (i.e., SAT scores and dropout rates) in the state based on school characteristics and household demographics. The school characteristics we used include per pupil expenditure in the state. Our household demographic measures included the share of blacks and Hispanics in the state population, the fraction of adults who are college educated, and median household income. For instance, if a state has a higher dropout rate than one would predict given its characteristics, we considered that state’s school system to be underperforming. Compared with parents and policymakers in states where expectations are being met or exceeded, those in states with underperforming schools may be more supportive of education reforms such as charter schools.
Of course, whether educational preferences based on demographics or dissatisfaction with existing school performance manifest themselves in support for charter schools depends on other circumstances as well: notably, the political power of opponents to charter schools, the most prominent opponents being teachers unions; and the degree of school choice already available to parents. We estimated the power of unions with the fraction of all teachers in a state who are union members. We based our assessment of the degree of school choice in a state on several indicators, including the fraction of students in the state enrolled in private schools and an index that measures the extent of choice parents have between school districts in the state as of the 1989–90 school year. The landscape of school choice options may be important in that demand for public charters could be greater where parents have fewer schooling options.
The demographic and political characteristics of a state and character of the state law authorizing charter schools undoubtedly matter in some way for the fate of charter schools in a state, but most decisions about charter school formation and attendance are made within school districts—by founders who decide to start a new school, by authorizers who empower them to do so, and, ultimately, by parents who decide to enroll their students. Therefore, we also gave our attention to two indicators of the spread of charter schooling. We measured expansion of charters across districts: first, by whether the district had at least one operating charter school in the 2003–04 school year and, second, by the fraction of public school students enrolled in charter schools during the 2003–04 school year.
We compared districts with at least one charter to districts with no charters and compared districts with higher and lower enrollments in charter schools to search for differences among districts that could explain the variation. As with our analysis of differences among states, we estimated the roles played by demographics, student achievement, and degree of school choice currently available in explaining variation in support for charter schools. We also accounted for whether the district is urban, suburban, or rural; whether the district is elementary, secondary, or unified; per pupil expenditure in the district during the 1989–90 school year; and differences attributable to the state. State-level differences included the strength of charter laws, statewide demographics, existing school choice policies, number of school districts, and the presence of charter support or opposition groups that operate throughout the state.
Explaining Support from State to State
State demographics, student academic performance (measured relative to expectations), and teachers union strength all play important roles in shaping state charter legislation and student participation. Let’s begin with demographics. States with larger Hispanic populations tended to pass laws supporting charter schools earlier and were likely to pass more-permissive legislation. For example, states with an Hispanic-population of 14 percentage points (two standard deviations) higher than the average were about 10 percent more likely to pass a charter law. These states also passed laws that ranked 12 to 14 points higher on the CER strength index (out of a total of 50). There is not, however, strong evidence that these states also had a greater proportion of their students enrolled in charter schools in 2003–04.
Considering the effects of Hispanic population on charter laws, we were surprised that the fraction of a state’s population that is black did not affect the likelihood that a state would pass a charter law, nor did we find any relationship to the timing of passage or the strength of the law. We were very interested to discover that the size of a state’s black population does, however, have a strong relationship with a state’s charter enrollment. For example, a 12.1-percentage-point increase (one standard deviation) in the fraction of a state’s population that is black is associated with roughly a 2-percentage-point increase in charter school enrollment in the state. This is effectively double the charter school enrollment in the average state. One potential explanation for this finding is that black voters, being heavily concentrated in the Democratic Party, are not traditionally swing voters who could influence state legislators’ positions on charter legislation. But once charters are established, African American families are active supporters of charter schools.
Strong charter laws appeared earlier in states where the fraction of adults with at least a college education was higher. For example, states in which the college-educated share of the population was 2.4 percentage points higher passed laws on average six months earlier. States with more-educated populations also had a greater fraction of students enrolled in charter schools. The size of this estimated effect is similar to the effect of a larger black population, about 2 percent greater enrollment in charter schools accompanying a one-standard-deviation increase in the fraction of college-educated adults in the state. Our data did not allow us to explain what lies behind this relationship, but we do have two conjectures. Highly educated citizens may have a greater willingness to experiment with education reforms, or there may be a “supply side” phenomenon: more educated adults translates into a larger pool of charter suppliers.
When we looked at changes in state demographics over time, we found that states with growing Hispanic and college-educated populations were more likely to pass early charter laws. We also found that states with growing income inequality during the 1980s were more likely to pass laws and to pass stronger charter laws during the 1990s.
States with higher-than-expected SAT scores were less likely to pass charter school legislation; tended to adopt such legislation later, if at all; and passed weaker laws. There is no evidence, however, of a statistically significant relationship between state SAT performance and enrollment in charter schools.
A higher- or lower-than-expected high-school dropout rate has no clear relationship to the passage or strength of charter legislation, but does, interestingly, have a strong relationship with charter school participation. States with higher-than-predicted dropout rates had significantly higher enrollment in charters. If the dropout rate is an additional 2 percentage points, or roughly one standard deviation, higher than expected, a state experiences a 1-percentage-point increase in charter school enrollment.
What could account for the differences in the estimated effects of SAT performance and the high-school dropout rates? It may be that SAT scores, as a very public measure of school performance, lead to agitation for charter laws, but that charters themselves are more likely to target students at risk of dropping out, and therefore participation is more closely associated with dropout rates.
Finally, membership in teachers unions has a substantial impact on the legal status of charter schools. States where a greater fraction of teachers were covered by a union contract in 1987 were much less likely to pass a charter law in the 1990s, more likely to pass a law later (if at all), and more likely to pass a weaker law. A one-standard-deviation increase in the fraction of teachers who are unionized, that is, an additional 20 percent of instructional employees covered by union contracts, means a state is 20 percent less likely to pass a charter law. We were initially puzzled to find that, conditional on the successful passage of a charter law and controlling for law strength, the fraction of students enrolled in charter schools appears to increase with the fraction of teachers in the state who are unionized. We share some further thoughts on these contradictory findings in our analysis of school districts.
We found little evidence that the extent of choice among districts in a state is related to either charter school legislation or participation, nor did we find a relationship between income or income inequality and legislative support for charter schools. We did find a positive relationship between the fraction of students enrolled in private schools prior to the passage of charter laws and law passage and strength. This may be due to private school parents supporting public charter schools as a substitute for private schools, or it may be related to broad dissatisfaction with public schools and a generally higher demand for alternatives.
Some have argued that the political composition of the governor’s office and the state legislature may influence the success of charter school laws. We reconducted all our statistical analysis, taking into account party control of the governor’s office and state legislature from 1990 to 2004. We never found party control of the state government to have an effect on the probability that a law passed, the year a law passed, the strength of the law, or participation in charter schools once we controlled for other state characteristics. In fact, of the 40 states passing a law, 24 passed it when a Republican was governor and 16 when a Democrat was governor, and about two-thirds of the states passing a charter law did so when there was no single party controlling both the legislature and the governor’s office.
School District Support for Charter Schools
We began to review the findings of our district analysis by first checking that the results from our statistical procedures were consistent with well-known patterns of enrollment in charter schools. As we expected, school districts with only secondary schools or both secondary and elementary schools were more likely to have a charter school in 2003–04 than districts with only elementary schools. School districts primarily based in a large or mid-sized city or large town were more likely to have a charter school than suburban or rural school districts.
We found that the fraction of a school district’s population that is Hispanic has no clear relationship with either the presence of charter schools or enrollment in the district. As in our state analysis, an increase in the fraction of a school district’s population that is black makes a district more likely to have a charter school in operation and to have a greater share of its students enrolled in charter schools. We estimated that an increase of 11 percentage points in the black population in a district increases the share of students enrolled in charter schools by about 6 percentage points. Considering that the average share of students enrolled in charter schools in school districts with a charter school is only about 10 percent, this is a substantial effect.
We also examined changes in demographic characteristics between 1980 and 1990, and found that districts with a rising fraction of black or college-educated individuals saw greater participation in charter schools. In addition, we found that districts where income inequality was rising saw higher participation in charter schools. This finding may reflect a divergence in preferences for education programs, for example, or an increasing diversity of needs within school districts that raise the demand for charter school options.
School districts that already had higher fractions of students enrolled in private schools, even accounting for the urban or rural location of the district, had a greater likelihood of having a charter school open in their district by 2003–04 and a greater share of their students enrolled in charters. Whether this pattern is indicative of general receptiveness on the part of these districts toward alternatives to public schools or a long-standing dissatisfaction with traditional public schools, it certainly suggests that private schools do not serve as a hindrance to the start-up of public charter schools. Our measure of the current level of choice in the public school system has no statistically significant relationship with charter support within school districts.
Districts with higher-than-predicted high-school dropout rates were, like states with high dropout rates, more likely to have charter schools and a greater share of students enrolled in charters. An 8-percentage-point increase in the adjusted dropout rate (that is, the average deviation from the predicted dropout rate) was associated with about a 2-percentage-point increase in the fraction of students enrolled in charters.
We were very interested to see the relationship between the strength of teachers unions and charter school enrollment, which we first glimpsed in our state analysis, reemerge in our study of districts. Contrary to what one might expect given the opposition—or at least hearty skepticism—of teachers unions to the charter school movement, districts with a greater union presence were more likely to have a charter school and to have a greater share of public school students enrolled in charter schools in 2003–04. A 39-percentage-point increase in the fraction of teachers unionized is associated with about a 2-percentage-point increase in the fraction of students enrolled in charters.
What lies behind this relationship between teachers’ unionization and support for charter schools? We cannot say for sure with our data, but it may be that parents are more likely to support charter schools in heavily unionized states, perhaps in a desire for more local control and less bureaucracy, or a desire for curricular and personnel policies that are less influenced by the union. Strong unions are more successful than weaker ones in opposing liberal charter legislation, but once a charter law is adopted, it seems that parents see charters as an avenue for reform in districts where unions have a strong hold on traditional public schools.
Understanding why states pass laws favorable to charters and why charter school participation is higher in some areas than others is important if we want to identify locations where school choice reforms are likely to present meaningful alternatives to traditional public schools and where they are likely to remain reforms on the margins. Our findings suggest that there are several forces propelling the expansion of the charter school movement. One of the most powerful is growing diversity in state and district populations. States and school districts with more blacks and college-educated adults have a substantially larger share of their students in charter schools than other districts. Failure to meet expectations for student academic performance, as measured by SAT scores and high-school dropout rates, also fuels the passage of charter laws, leads to the passage of stronger charter laws, contributes to the creation of charter schools, and boosts charter school enrollment. Unexpectedly, we discovered that a leading opponent of the charter school movement, teachers unions, appears to contribute indirectly to the expansion of charter schools. In states where teachers are unionized at the highest rates, charter laws were less likely to be enacted, were passed later, and were less favorable to charter schools in states that did adopt charter laws. But in states and school districts with strong unions and charter laws of similar strength, more families have sought out alternatives for their children in charter schools. All else being equal, a highly unionized teaching labor force goes hand in hand with the creation of more charter schools and more students learning in charter school classrooms.
-Christiana Stoddard is assistant professor in the department of agricultural economics and economics at Montana State University. Sean Corcoran is assistant professor of educational economics at the Steinhardt School of Education, New York University.