By Justin Hill, Johns Hopkins University
A recent survey experiment conducted in New York City by Chantal A. Hailey provides insight on the influence of school racial composition during the school selection process for parents and students. The author conducted this survey at school fairs for eighth-grade students and their parents and asked respondents to evaluate five randomly-selected hypothetical high school profiles. The profiles included a majority White school, majority Black school, majority Latinx school, and a racially diverse school. The profiles also had random variation in safety rating, graduation rate, and metal detector presence. For each school, students and parents were asked to indicate their likelihood of choosing each school on a 7-point Likert scale, with 1 indicating very unlikely and 7 indicating very likely. The sample included 403 parents and 431 students whose racial identities were similar to the New York City population and whose socioeconomic profiles were slightly higher than the New York City student population.
In comparison to the majority White school, White respondents were less likely to choose the racially diverse school (b = -0.44, p < .01), the majority Latinx school (b = -0.75, p < .001), and the majority Black school (b = -0.90, p < .001), after controlling for common covariates and other experimental conditions. Also of interest is that the magnitude of White respondents’ difference in preference between White majority and Black majority schools was over two times larger than the effect of graduation rate (b = 0.42, p < .001) on school selection for White respondents. In comparison to the majority White school, Asian respondents were also less likely to choose the majority Latinx school (b = -0.74, p < .001) and the majority Black school (b = -0.65, p < .001). However, Asian respondents did not show a significant difference in preference for the majority White school when compared to the racially diverse school. In comparison to the majority Latinx school, Latinx respondents were less likely to choose the majority White school (b = -0.22, p < .05) and the majority Black school (b = -0.41, p < .001), but did not show a significant difference in preference for the racially diverse school. In comparison to the majority Black school, Black respondents were less likely to choose the majority White school (b = -0.33, p < .01), but did not show a significant difference in preference for the majority Latinx school or the racially diverse school.
The results indicate that White and Asian respondents hold similar school preferences that are different from Latinx or Black respondents. The author also attempted to provide verification for the results of the analysis by reviewing 2013 administrative data on school selection, finding that the results of the survey experiment aligned closely with school choices evident from the administrative data. While it is possible that respondents’ stereotypes about the relative economic advantage or disadvantage of schools with certain racial demographics played a role in their decision-making, the results still suggest that school-choice policies within districts may exacerbate racial segregation rather than help create more racially diverse schools.
Best Evidence in Brief Index 