Association for Public Policy Analysis & Management

Comparing student sorting and mobility patterns across the school choice programs of charter and magnet schools.

Ron Zimmer

Adam Kho

Vanderbilt University

An important question for the school choice movement is the effect these programs have on the sorting of students. Two of the most prominent modes of school choice are magnet and charter schools. While magnet programs were established in the 1960s and 1970s as means of creating greater racial integration, charter schools were promoted primarily as means of creating innovation and improving student achievement. Nevertheless, there are concerns as to whether charter schools may further stratify an already racially stratified system as well create “churns” of students moving in and out of schools or “skim the cream,” enrolling high-ability students at the expense of lower achievers left in traditional public schools (TPS). With this in mind, an emerging literature has examined the effect charter schools have on student sorting. However, while some of these same concerns could be expressed for magnet schools, little research has examined student sorting for these schools.

For charter schools, the emerging research has mostly focused on racial segregation in which researchers have generally compared the racial makeup of charter schools relative to state and district averages, not taking into account the fact that charter schools are not randomly dispersed within a state or district. To gain a better sense of whether a student is in a more segregated environment, a few studies have taken the additional step using longitudinal student-level data to compare the racial makeup of the TPS a student transfers out of to the charter school the student transfers into (Booker, et al., 2005; Bifulco and Ladd, 2007; Zimmer et al., 2009). A subset of these studies have also provided insights into whether a charter school is cream skimming high ability students by examining whether students transferring out of a TPS and into a charter school were above or below the ability level of the TPS’s average student, but this research only examined student transfers across years, not within years. Similarly, two studies (Zimmer and Guarino, 2014; Winters, 2015) examined the achievement level of students transferring out of charter schools to see if there is evidence of schools pushing out low-performing students, but again, these studies do not have access to within school year moves. In addition, no studies have examined where low-performing students go when they exit a school of choice (e.g., do they go to TPSs or another school of choice) and whether schools of choice are less likely to fill the spot of a low-performing student that exits than a TPS (i.e., less likely to backfill).    

In this paper, we examine both charter and magnet schools to see whether these school choice policies are leading to increased racial segregation, cream skimming, high exit rates of low-performing students, greater level of student churn, or less “backfilling”. To carry out these analyses, we use statewide Tennessee longitudinal data that includes both within and across school year student moves.