Search and Filter

Submit a paper

Not yet affiliated? Have a paper you wish to post? Check out the EdWorkingPapers' scope and FAQs, and then submit your manuscript here.

A Descriptive Analysis of Cream Skimming and Pushout in Choice versus Traditional Public Schools

One of the controversies surrounding charter schools is whether these schools may either “cream skim” high-performing students from traditional public schools or “pushout” low-achieving students or students with discipline histories, leaving traditional public schools to educate the most challenging students. We use these terms strictly for brevity and acknowledge that many of the reasons that students are labeled high- or low-performing academically or behaviorally are beyond the control of the student. In this study, we use longitudinal statewide data from Tennessee and North Carolina and linear probability models to examine whether there is evidence consistent with these selective enrollment practices. Because school choice programs managed by districts (magnet and open enrollment programs) have a similar ability to cream skim and pushout students, we also examine these outcomes for these programs. Across the various school choice programs, magnet schools have the most evidence of cream skimming, but this might be expected as they often have selective admissions. For charter schools, we do not find patterns in the data consistent with cream skimming, but we do find evidence consistent with pushout behaviors based on discipline records. Finally, some have raised concerns that students may be pushed out near accountability test dates, but our results suggest no evidence consistent with this claim.

Keywords
school choice, inequality, charter schools
Education level
Topics
Document Object Identifier (DOI)
10.26300/2wns-aq26
EdWorkingPaper suggested citation:
Kho, Adam, Ron Zimmer, and Andrew McEachin. (). A Descriptive Analysis of Cream Skimming and Pushout in Choice versus Traditional Public Schools. (EdWorkingPaper: -332). Retrieved from Annenberg Institute at Brown University: https://doi.org/10.26300/2wns-aq26

Machine-readable bibliographic record: RIS, BibTeX