What Two Decades of Research Tells Us About School Choice

Spotlight on Competition, Accountability Standards and Charter Schools

School bus
Jonathan Macagba / Getty Images

The concept of school choice as we know it today has been around since the 1950s when economist Milton Friedman began making arguments for school vouchers. Friedman argued, from an economics standpoint, that education should, in fact, be funded by the government, but that parents should have the freedom to choose whether their child would attend private or public school.

Today, school choice encompasses several options in addition to vouchers, including neighborhood public schools, magnet schools, charter public schools, tuition tax credits, homeschooling, and supplemental educational services. More than half a century after Friedman articulated the still popular economist's argument for school choice, 31 U.S. states offer some form of school choice program, according to EdChoice, a non-profit organization that supports school choice initiatives and was founded by Friedman and his wife, Rose.

Data show that these changes have come swiftly. According to The Washington Post, just three decades ago there were no state voucher programs. But now, per EdChoice, 29 states offer them and have diverted 400,000 students to private schools. Similarly and even more striking, the first charter school opened in 1992, and just a little more than two decades later, there were 6,400 charter schools serving 2.5 million students across the U.S. in 2014, according to sociologist Mark Berends.

Common Arguments For and Against School Choice

The argument in support of school choice uses economic logic to suggest that giving parents a choice in which schools their children attend creates healthy competition among schools. Economists believe that improvements in products and services follow competition, so, they reason that competition among schools raises the quality of education for all. Advocates point to historical and contemporary unequal access to education as another reason to support school choice programs that free children from poor or struggling zip codes and allow them to attend better schools in other areas. Many make racial justice claims about this aspect of school choice since it is primarily racial minority students who are clustered in struggling and underfunded schools.

These arguments seem to hold sway. According to a 2016 survey conducted by EdChoice, there is overwhelming support among state legislators for school choice programs, especially educational savings accounts and charter schools. In fact, school choice programs are so widely popular among legislators that it is a rare bipartisan issue in today's political landscape. President Obama's education policy championed and provided massive amounts of funding for charter schools, and President Trump and Secretary of Education Betsy DeVos are vocal supporters of these and other school choice initiatives.

But critics, notably teachers unions, claim that school choice programs divert much-needed funding away from public schools, thus undermining the public education system. In particular, they point out that school voucher programs allow taxpayer dollars to go to private and religious schools. They argue that, instead, in order for high-quality education to be available to all, regardless of race or class, the public system must be protected, supported, and improved. Still, others point out that there is no empirical evidence to support the economics argument that school choice fosters productive competition among schools.

Passionate and logical arguments are made on both sides, but in order to understand which should hold sway over policymakers, it's necessary to look at the social science research on school choice programs to determine which arguments are more sound.

Increased State Funding, Not Competition, Improves Public Schools

The argument that competition among schools improves the quality of the education they provide is a long-standing one that is used to support arguments for school choice initiatives, but is there any evidence that it is true? Sociologist Richard Arum set out to examine the validity of this theory way back in 1996 when school choice meant choosing between public and private schools. Specifically, he wanted to know whether competition from private schools impacts the organizational structure of public schools, and if, in doing so, competition has an impact on student outcomes. Arum used statistical analysis to study the relationships between the size of the private school sector in a given state and the scope of public school resources measured as student/teacher ratio, and the relationship between student/teacher ratio in a given state and student outcomes as measured by performance on standardized tests.

The results of Arum's study, published in American Sociological Review, the top-ranking journal in the field, show that the presence of private schools does not make public schools better through market pressure. Rather, states in which there are high numbers of private schools invest more finances in public education than do others, and so, their students do better on standardized tests. Notably, his study found that spending per student in a given state increased significantly along with the size of the private school sector, and it is this increased spending that leads to lower student/teacher ratios. Ultimately, Arum concluded that it was increased funding at the school level that led to better student outcomes, rather than a direct effect of competition from the private school sector. So while it is true that competition among private and public schools can lead to improved outcomes, competition itself is not enough to foster those improvements. Improvements only occur when states invest heightened resources in their public schools.

What We Think We Know about Failing Schools is Wrong

A key part of the logic of arguments for school choice is that parents should have the right to pull their children out of low-performing or failing schools and send them instead to schools that perform better. Within the U.S., how school performance is measured is with standardized test scores meant to indicate student achievement, so whether or not a school is considered to be successful or failing at educating students is based on how students at that school score. By this measure, schools whose students score in the bottom twenty percent of all students are considered to be failing. Based on this measure of achievement, some failing schools are shut down, and, in some cases, replaced by charter schools.

However, many educators and social scientists who study education believe that standardized tests are not necessarily an accurate measure of how much students learn in a given school year. Critics point out that such tests measure students on just one day of the year and do not account for external factors or differences in learning that might influence student performance. In 2008, sociologists Douglas B. Downey, Paul T. von Hippel, Melanie Hughes decided to study just how different student test scores might be from learning outcomes as measured by other means, and how different measures might impact whether or not a school is classified as failing.

To examine student outcomes differently, the researchers measured learning by evaluating how much students learned in a given year. They did this by relying on data from the Early Childhood Longitudinal Study conducted by the National Center for Education Statistics, which tracked a cohort of children from kindergarten in the fall of 1998 through the end of their fifth-grade year in 2004. Using a sample of 4,217 children from 287 schools across the country, Downey and his team zoomed in on the change in performance on tests for the children from the beginning of kindergarten through the fall of first grade. In addition, they measured the impact of the school by looking at the difference between the learning rates of students in first grade versus their learning rate during the previous summer.

What they found was shocking. Using these measures, Downey and colleagues revealed that less than half of all schools that are classified as failing according to test scores are considered as failing when measured by student learning or educational impact. What's more, they found that about 20 percent of schools "with satisfactory achievement scores turn up among the poorest performers with respect to learning or impact."

In the report, the researchers point out that most of the schools that are failing in terms of achievement are public schools that serve poor and racial minority students in urban areas. Because of this, some people believe that the public school system is simply unable to adequately serve these communities, or that children from this sector of society are unteachable. But the results of Downey's study show that when measured for learning, the socioeconomic differences between failing and successful schools either shrink or disappear entirely. In terms of kindergarten and first-grade learning, the research shows that schools that rank in the bottom 20 percent "are not significantly more likely to be urban or public" than the rest. In terms of learning impact, the study found that the bottom 20 percent of schools are still more likely to have poor and minority students, but the differences between these schools and those that rank higher are considerably smaller than the difference between those that rank low and high for achievement.

The researchers conclude “when schools are evaluated with respect to achievement, schools that serve disadvantaged students are disproportionately likely to be labeled as failing. When schools are evaluated in terms of learning or impact, however, school failure appears to be less concentrated among disadvantaged groups.”

Charter Schools Have Mixed Results on Student Achievement

Over the last two decades, charter schools have become a staple of education reform and school choice initiatives. Their proponents champion them as incubators of innovative approaches to education and teaching, for having high academic standards that encourage students to reach their full potential, and as an important source of educational choice for Black, Latino, and Hispanic families, whose children are disproportionately served by charters. But do they actually live up to the hype and do a better job than public schools?

To answer this question, sociologist Mark Berends conducted a systematic review of all published, peer-reviewed studies of charter schools conducted over twenty years. He found that the studies show that while there are some examples of success, particularly in large urban school districts that primarily serve students of color like those in New York City and Boston, they also show that across the nation, there is little evidence that charters do better than traditional public schools when it comes to student test scores.

The study conducted by Berends, and published in the Annual Review of Sociology in 2015, explains that in both New York and Boston, researchers found that students attending charter schools closed or significantly narrowed what is known as "the racial achievement gap" in both mathematics and English/language arts, as measured by standardized test scores. Another study Berends reviewed found that students who attended charter schools in Florida were more likely to graduate high school, enroll in college and study for at least two years, and earn more money than their peers who did not attend charters. However, he cautions that findings like these appear to be particular to urban areas where school reforms have been difficult to pass.

Other studies of charter schools from across the country, however, find either no gains or mixed outcomes in terms of student performance on standardized tests. Perhaps this is because Berends also found that charter schools, in how they actually operate, are not so different from successful public schools. While charter schools might be innovative in terms of organizational structure, studies from around the country show that the characteristics that make charter schools effective are the same ones that make public schools effective. Further, the research shows that when looking at practices within the classroom, there is little difference between charters and public schools.

Taking all of this research into consideration, it seems that school choice reforms should be approached with a healthy amount of skepticism as to their stated goals and intended outcomes.