The National Education Policy Center has reprinted a methodical, well footnoted analysis by Kevin Welner documenting the (often subtle) ways in which charters get the kids the want and avoid the kids they don’t want. This happens more often the national networks rather than the locally operated charters we have here in New Hampshire but this is worth having at hand as legislators review state charter policy and potentially recommend new rules for charter applications and governance.
by Kevin G. Welner — April 22, 2013
This commentary offers a classification of twelve different approaches that charter schools use to structure their student enrollment. These practices impact the likelihood of students enrolling with a given set of characteristics, be it higher (or lower) test scores, students with ‘expensive’ disabilities, English learners, students of color, or students in poverty.
Informed discussions of whether charter schools are successful include considerations of student differences between different schools. We address “selection bias,” “attrition bias” and the like because measured outcomes are undoubtedly affected by inputs. Charter schools tend to have fewer students with disabilities, fewer English learners, and a less poor population of students than their surrounding public schools (Frankenberg, Siegel-Hawley, and Wang, 2010; Miron, et al., 2010; U.S. Government Accountability Office, 2012).
In fact, the patterns are particularly stark when we realize that such at-risk students are disproportionately enrolled in a small subset of “mission-oriented” charters – those dedicated to serving a particular type of at-risk student. For instance, the “majority of charter school students with severe disabilities [in Florida] are concentrated in a handful of schools that specialize in those disabilities….” (O’Connor, J., & Gonzalez, 2011; see also Miron, et al.  for a national picture). This leaves the remaining charters serving even fewer at-risk children.