by Eric A. Hanushek, John F. Kain, Jacob M. Markman, and Steven G. Rivkin *
Abstract Empirical analysis of peer effects on student achievement has been limited, and what exists has been open to question because of the difficulties of identifying peer effects per se. Our strategy to identify peer group effects involves the elimination of problems introduced by unmeasured or mismeasured influences on achievement through the estimation of achievement growth models with fixed individual and school-by-grade effects. Our basic estimation of elementary school achievement growth indicates that the achievement level and racial composition of peers has a direct influence on achievement. All students appear to benefit from having higher achieving schoolmates, although the effect is quite small. The variance in achievement appears to have no systematic influence, and the effects of mean differences in peer achievement levels are roughly constant across quartiles of the achievement distribution – suggesting that ability grouping policies have primary influence on the distribution of performance and not the level. Moreover, ceteris paribus schools with higher concentrations of minority students lead to lower achievement for Black students but minimal effects on whites or Hispanics.
Paper prepared for the Conference on Empirics of Social Interactions Brookings Institution January 14-15, 2000
University of Rochester and National Bureau of Economic Research; University of Texas at Dallas; Amherst College; and Amherst College, respectively. Support for this work has been provided by the Spencer Foundation, the Mellon Foundation, the Smith Richardson Foundation, and the Packard Humanities Institute. *
Do Peers Affect Student Achievement?
by Eric A. Hanushek, John F. Kain, Jacob M. Markman, and Steven G. Rivkin The peer group composition of schools is undeniably important in the minds of parents as well as policy makers at the local, state, and federal level. Residential location decisions of families, various state and federal laws, and court interpretations of school district policies appear to have an implicit if not explicit peer group component. There have nonetheless been relatively few direct investigations of the impact of peer groups on student performance, and what evidence exists is open to widely varying interpretations. This paper uses a unique identification strategy to extract the causal effects of peer characteristics on student achievement and thus to assess the quantitative potential of alternative policy interventions. Perhaps no issue has engendered more passionate discussion about peer influences than school desegregation. Almost one half century after the landmark decision in Brown v. Board of Education, debate persists over the contribution of school segregation to the lower levels of academic and labor market success of African Americans. While subsequent court deliberations have considered a variety of factors including school resource distributions, the racial composition of schools continues to be a key element. The original Brown decision ruled that the doctrine of separate but equal was unconstitutional, because separate schools were inherently unequal. This decision was based partly on small-scale psychological evidence about the harm to Black students of racially isolated environments, though it subsequently received support from the influential Coleman Report (1966) and its offshoots (U. S. Commission on Civil Rights 1967). Nonetheless, there remains considerable disagreement about the nature and magnitude of benefits of various policy actions, let alone about the costs of these (e.g., Crane and Mahard 1978; Armor 1995). Peer group effects also have played a prominent role in a number of other policy debates including ability tracking and anti-poverty policies in both rural areas and urban ghettos. In
addition, opposition to the growing...