What Do We Know about School Effectiveness?

Abstract

Critics have challenged previous studies' use of NAEP data to refute the widely held belief that private schools are inherently better than public schools. To answer the critics, the authors conducted a new study, based on a rich longitudinal dataset. The results deliver another blow to the assumption of private school superiority. IN 2005, the Kappan published a report of our research on student achievement in public and private schools, based on an analysis of the 2000 National Assessment of Educational Progress (NAEP). Like most people, we had assumed that the higher average scores in private schools meant that private schools were more effective--an assumption that undergirds much of the current thinking surrounding education policies and reforms. But to our surprise, the data on a nationally representative sample of 30,000 students in fourth and eighth grades showed public schools to be outperforming private schools in mathematics achievement after student background factors were considered. (1) We subsequently examined this issue with the even more comprehensive 2003 NAEP data, covering a representative sample of almost 345,000 students. We found similar patterns, with public schools outperforming private schools and charter schools once we accounted for demographic differences in the populations they served. (2) Later, our results were confirmed in a report published by the U.S. Department of Education. (3) However, other researchers and some policy advocates then weighed in on this new school debate, challenging the data and methodologies used to address this question and calling for longitudinal examinations of public and private school effectiveness. (4) Our purpose here is twofold. First, we briefly outline the renewed debate on public and private school achievement, considering the state of current knowledge, political schisms, and the implications for current policies and proposals. Second, we report on a new study, not of student achievement at one point in time, but on student gains over time in public and private schools, using data from the Early Childhood Longitudinal Study, Kindergarten Class of 1998-99 (ECLS-K). The results of this study are both intriguing and illuminating for the crucial question of the effectiveness of public and private schools. THE CONTENTIOUS TERRAIN OF RESEARCH ON PUBLIC AND PRIVATE SCHOOLS Until the Kappan piece was published, the specific question of achievement in public and private schools had been relatively dormant for many years. Yet the issues underlying that question have been very much at the forefront of public policy discussions. Although generations of parents chose private schools largely for religious reasons, with the emergence of proposals for voucher programs that provide state subsidies for parents to enroll their children in private schools, policy makers and economists became more interested in achievement in different types of schools. Of course, researchers, parents, and policy makers have known for a long time that private schools tend to post higher test scores, on average, than public schools. But it was also recognized that private schools, on average, serve more advantaged students--those who already exhibit characteristics associated with school success. So the question was whether private schools were, themselves, causing greater gains in student achievement or whether their scores just reflected their more advantaged student population. If it turned out to be the former, this would suggest that distinctive policies, processes, and practices in private schools were shaping such gains and that either those attributes needed to be replicated in public schools or, if that were not possible (as many theorists have argued), then students should be encouraged to attend private schools. (5) For the last two decades or so, most people (ourselves included) had assumed that the public/private question was answered. …