Skip to Content


Virginia Journal of Education

Are Private Schools Superior to Public Ones?

University of Virginia researchers say private schools do no better at closing achievement gaps than public ones do.

By Robert Pianta and Arya Ansari

One of the most highly-touted reforms in our public schools, for many, has been the opportunity for low-income and other vulnerable children to use vouchers or other means to attend a private school. Supporters back such programs with the assumption that private schools are more effective, that they produce higher achievement, foster positive social adjustment and citizenship, and decrease risky behavior. 

To learn more and to examine the accuracy of such thinking, we did a unique longitudinal study of a large and diverse sample of children to find out how much being in private schools was an accurate predictor of achievement, social, and personal outcomes at age 15.

Some background for our research: First, studies show overwhelming evidence that factors such as income and parent education drive enrollment in private schools. Higher income families can afford private schools and choose ones that match goals for their children. Further, family factors (income, education, expectations) are also important in the stimulation and opportunity that drive children’s learning and achievement. In our study, we wanted to separate the effects of private schooling from the family background factors that play such a large role in children’s achievement. 

A second consideration is the exceptionally wide variation in private schooling, reflecting religious denominations, specialized curricula (e.g., Montessori, Waldorf), and how they’re run, ranging from institutionalized (e.g. the Archdiocese of New York) to localized. Leadership, staff qualifications, and school size also vary among private schools. Thus, it is important to keep in mind the significant differences in such schools, which is often seen as desirable by reformers, advocates, and parents, but can also lead to highly variable results with students.

Third, most well-controlled evaluations of voucher programs or private school enrollment tend to use state math and language arts standardized test scores. Such results don’t reflect other desirable student outcomes, such as motivation, social adjustment, and behavior, often given as reasons for private school enrollment, which have not been addressed in research.

In our study, we used the NICHD* Study of Early Child Care and Youth Development (SECCYD). SECCYD is a unique opportunity to study the effects of private schools because of its more comprehensive assessment of student outcomes, time frame longer than typical studies of private schooling, and detailed and wide-ranging assessment of family background and context. Its sample also reflects a broad range of the economic conditions, cultural beliefs, and child-rearing practices of the U.S. 

In our initial analyses, based on test scores alone, children with a history of enrollment in private schools did better than children in public schools. However, those differences were completely eliminated when we took into account family income.

We did further analyses to test the idea that private schooling might have a particular benefit for poor children, and found no evidence to support it.

We studied not only achievement, the sole focus of evaluations of private schooling reported to date, but also students’ social adjustment, attitudes, and motivation, and even risky behavior, all of which one assumes might be influenced by a private school education, and found no impacts. In short, despite the pronounced arguments in favor of policies and finances to support enrollment in private schools as a solution for vulnerable children, our study found no evidence that private schools, when family background (particularly income) is considered, are more effective for promoting student success than public ones.

While some studies have suggested that if a low-income student stays in a private school for at least four years, he or she increases the likelihood of gaining substantial benefits, we found that length of enrollment (on average 6 years) did not predict student outcomes, again, once family income was considered. Thus, even for students who remained in private school for almost half of their K-12 experience, there was no discernable effect on the wide range of outcomes we assessed at age 15.

Our study was not a purely experimental one in which students were enrolled in public or private school at random, thus eliminating family background factors as explanations for any benefits. Instead, we used statistical models to adjust for family, student, and contextual variables. And because our SECCYD sample was recruited from 10 sites across the country, reflecting more than 600 public and private schools and numerous school systems, differences among those schools might also reduce the likelihood of finding differences. 

To sum up our findings, we found no evidence to support the idea that widespread enrollment in private schools is a solution for achievement gaps associated with income or race.

When educational opportunities are discussed, people often assume that poor children attend low-quality schools and that their families, given resources and flexibility, would choose among a supply of private schools for their children because those schools are more effective. Our results do not show that this logic holds in the real world of a limited supply of effective schools, both private and public, or that once you account for family background, private schools result in a superior education. Actually, several studies suggest that to the extent that schooling benefits poor children, it results from enrollment in public schools, which are subject to greater oversight and more explicitly articulated standards for achievement than are private schools. It seems most important to better understand and use the mechanisms, in schools of any type and in families, that support student success and to do all we can to strengthen those resources.

*National Institute of Child Health and Human Development

Dr. Pianta is dean of the University of Virginia Curry School of Education and Human Development. Dr. Ansari is a postdoctoral research associate there.



Take action to boost K-12 funding and support better pay.


Stay in touch with VEA and your fellow members.

Check out VEA and NEA Member Benefits savings programs.

Embed This Page (x)

Select and copy this code to your clipboard