Ruth Curran Neild, University of Pennsylvania Graduate School of Education
April 11, 2003
Do magnet schools "cream" students from other local schools, harming those left behind? That's the focus of this paper, which examines 1996 data from Philadelphia, which initiated a magnet system in the 1970s in response to desegregation mandates. Neild finds that neighborhood schools lose, on average, 10 percent of their students to magnet schools. There is great variation, though and, because magnets do take good students, it's actually the best of the neighborhood schools that lose out: one-quarter of all neighborhood schools are "significantly" affected, and all of these are medium- to high-achieving schools. Of the least-affected schools, most are struggling. As a result, the best schools take the largest hit from magnets even though they still remain at the top of the achievement heap. Interestingly, Neild found similar results with lottery magnets (those with random admissions policies); the best schools were affected much more than the worst schools. Such results suggest that those hoping to improve public schools need not fight magnet schools; their impact on struggling schools is minimal, at best. The report is interesting though brief, and does not examine some other potential impacts of magnets - on competition among schools, on improving student achievement, or on parent satisfaction, to name a few. You can find a copy at http://www.gse.upenn.edu/~rneild/Neighborhood high schools.Neild.doc.
