Why racial segregation is more severe in public schools
Three weeks ago, we directed readers to an article in The Wall Street Journal by Jay Greene arguing that, contrary to what was reported by the Harvard Civil Rights Project, private schools are actually more integrated than public schools. ("Choosing Integration," July 8, 2002, available to subscribers only) The Wall Street Journal published a letter by Sean Reardon in response to Jay's piece. ("Vouchers, Private Schools and Segregation," July 25, 2002, available to subscribers only) In the interest of clarifying the debate over private schools and segregation, we present here a letter from Jay Greene responding to points made in Reardon's letter.
To the Editor,
Sean Reardon's July 25th letter in response to my July 8th op-ed did not offer a defense of his research findings on the effects of private education on racial integration; it merely repeated the same methodologically flawed results that my op-ed critiqued. In my op-ed, I described how the study by Mr. Reardon and John Yun produced statistically biased results because it over-represented private elementary schools. This is significant because elementary schools - public or private - tend to be more segregated because they usually draw students from smaller neighborhoods. Reardon and Yun's results really only show that private schools educate younger students, not that they are more segregated. Their findings are also distorted by looking at school-level data that are misleading because of the fairly common public school practice of re-segregating students within schools by tracking, racially-biased course assignments, and housing all- or mostly-white magnet schools in buildings located within minority neighborhoods.
My own research, looking at integration in the classroom and comparing students in public and private schools in the same grade (a national sample of 12th graders), produces results that are exactly the opposite of Reardon and Yun's claims. Without their methodological flaws, we find that private schools are significantly better integrated than public schools. Other than dismissing my critiques as "quibbling," Mr. Reardon offers no defense of his research methodology and does not explain why readers should not be more strongly persuaded by my positive findings about the effects of private education on classroom (and lunchroom) racial integration.
Mr. Reardon does respond to my analysis of the effects of Cleveland's voucher program on integration, arguing that it is inappropriate to compare integration in private schools receiving voucher students to public schools in the Cleveland metropolitan area because "families in Cleveland cannot choose from among the public schools in the suburbs of Cleveland..." But of course families can choose suburban public schools if they have the resources to move to the suburbs - residential choice being the most common (and least egalitarian) type of school choice. In fact, the greater ability of white families to exercise residential choice and leave the city's failing schools is precisely why racial integration is so severe around Cleveland. A family wishing to live and work in that area could choose a public school in the city or suburbs or (if poor) could choose to live in the city and receive a voucher to attend a private school. If they choose the voucher option, they have better odds of being in a racially mixed school. The fact that residential choice and the voucher program are both options makes the comparison of the voucher schools to all Cleveland metropolitan schools necessary and appropriate.
Mr. Reardon's theory for why segregation is more severe in private schools is something out of Alice's Wonderland. He contends that "private schools tend to be more racially segregated...because most private school students attend religious private schools in their local neighborhood... mirror[ing] the high levels of residential segregation in the U.S." Mr. Reardon has it exactly backward. Racial segregation in public schools is generally more severe because public school students are constrained by attendance zone and district boundaries to attend schools that reflect racially segregated housing patterns. Because private schools do not face these constraints, they have a better chance of mixing students from different neighborhoods and school districts. By expanding the ability of families to choose private schools and thereby further detaching schooling from housing, vouchers offer a promising avenue for reducing school segregation.
Jay P. Greene, Ph.D.
Manhattan Institute for Policy Research