This website uses cookies.

By clicking the "Accept" button or continuing to browse our site, you agree to first-party and session-only cookies being stored on your device to enhance site navigation and analyze site performance and traffic. For more information on our use of cookies, please see our Privacy Policy.

  1. Home
  2. Journals
  3. American Economic Journal: Economic Policy
  4. May 2020
  5. Does School Desegregation Promote Diverse Interactions? An Equilibrium Model of Segregation within Schools

Does School Desegregation Promote Diverse Interactions? An Equilibrium Model of Segregation within Schools

  • Angelo Mele
American Economic Journal: Economic Policy
vol. 12, no. 2, May 2020
(pp. 228-57)
Download Full Text PDF

Abstract

This paper studies racial segregation in schools using data on student friendships from Add Health. I estimate an equilibrium model of friendship formation, with preferences allowing both homophily and heterophily in direct and indirect ties. I find that homophily goes beyond direct links: students also prefer racially homogeneous indirect friends, while there is heterophily in income. I simulate policies reallocating students across schools. Race-based policies have nonlinear effects on within-school segregation and other network features such as clustering and centrality. Policies increasing diversity through reallocations based on income have less impact on racial segregation.

Citation

Mele, Angelo. 2020. "Does School Desegregation Promote Diverse Interactions? An Equilibrium Model of Segregation within Schools." American Economic Journal: Economic Policy, 12 (2): 228-57. DOI: 10.1257/pol.20170604

Additional Materials

JEL Classification

  • H75 State and Local Government: Health; Education; Welfare; Public Pensions
  • I21 Analysis of Education
  • I28 Education: Government Policy
  • J15 Economics of Minorities, Races, Indigenous Peoples, and Immigrants; Non-labor Discrimination

Missing Friends

  • Posted by Robert Kaestner
  • Posted on 4/30/2020 10:15:26 AM
The article is based on friendship information in Wave I (1994-95) in the National Longitudinal Study of Adolescent Health (Add Health). The sample includes students form 16 schools. The key information is about friendship links and the characteristics of those friendship networks.

However, the article does not report the number of students who report no friendships and friends that cannot be matched (e.g., outside school) to information necessary for the analysis. The number of students with no friends (is this a valid observation?), missing friends and non-matched friends is non-trivial (Fletcher JM, Ross SL. Estimating the effects of friends on health behaviors of adolescents. Health Econ. 2018;27(10):1450‐1483. doi:10.1002/hec.3780).

While the article makes both theoretical and empirical contributions, the potential for measurement error regarding friendship networks, which is unlikely to be random, may diminish the empirical contribution. The focus on 16 schools in 1994-95 also limits the external validity of the policy simulations.

I hope the author will provide information about the measurement issue I raise and whether it is likely to be important. Perhaps it is not.



, which is likely to be substantial
Login to Comment