Erika Frankenberg and Chungmei Lee, The Civil Rights Project, Harvard University
July 2003
When the Harvard Civil Rights Project issues a new study on school segregation, you really don't need to read it to know what it will say. Their recent report on segregation in charter schools is utterly predictable: according to the authors, charter schools are more segregated than their conventional counterparts. What's interesting is that their methodology is growing increasingly suspect. This time, to prove that charter schools are more segregated, they've compared charters' demographics with those of traditional public schools in the entire state. In other words, rather than comparing a charter with public schools in the same school district (which are more likely to have demographics similar to the local charters and which are also the plausible alternative option for students enrolled in charters), they compare the racial composition of a charter school to the demographics of all public schools in its state, including schools in more diverse neighborhoods, in affluent suburbs, in rural areas, etc. The authors reason that "because charter schools are created under state law and are, or could easily be [emphasis added] made, independent of district boundaries & it seems appropriate to compare them with other schools in the state rather than schools in the particular community where they are physically located." What a load of nonsense - and more proof of the old adage about lies and statistics. Should you care to subject yourself to this report, hold your nose and go to http://www.civilrightsproject.harvard.edu/research/deseg/Charter_Schools03.pdf.
