2007 - Unraveling Black Peer Effects on Black Test Scores

Attribution: Armor, David J., & Duck, Stephanie
Researchers: David J. ArmorStephanie Duck
University Affiliation: George Mason University
Email: darmor@gmu.edu
Research Question:
Replicate the Hanushek, Kain and Rivkin model using North Carolina state testing data
Published: 0
Journal Name or Institutional Affiliation: N/A
Journal Entry: Presented at the APPAM Research Conference
Year: 2007
Findings:
  • The finding of Black peer effects for White students does not replicate the Texas results.
  • The relationship between Black peers and achievement in North Carolina is quite different than that found in Texas. The Black peer effect in for Black students in NC schools are much smaller than those in Texas, and in many models the Black peer effect is small and not statistically significant. In the case of White students, the North Carolina analysis shows larger Black peer effects than in the Texas data, in some cases equaling or even surpassing the peer effects for Black students.
  • The NC analysis finds Black peer effects for White students, and in mathematics these effects are either equal to or stronger for Whites than Blacks.
  • It does not appear that North Carolina black students who spend most of their elementary and middle school years in predominantly Black schools are significantly harmed -in terms of academic achievement- as compared to Black students in integrated schools.
Keywords: Academic AchievementPeer EffectsRacial CompositionRegions: SouthMethodologies: QuantitativeResearch Designs: Secondary Survey DataAnalysis Methods: Fixed Effects Regression Models Sampling Frame:North Carolina Students
Sampling Types: NonrandomAnalysis Units: StudentData Types: Quantitative-Longitudinal
Data Description:
  • Uses North Carolina state testing data for grades 3 to 8 and from 1997 to 2005.
  • Includes four cohorts of students who attended grades 3 to 8 between 1997 and 2005. The first cohort attended grade 3 in 1997 and grade 8 in 2002: the last cohort attended grade 3 in 2000 and grade 8 in 2005.
  • Uses the twice-lagged test score variable as the instrument in the two-least squares regression.
  • Two basic forms of the analysis: using aggregate data and one using individual student data.
Theoretical Framework:
Relevance:
Archives: K-12 Integration, Desegregation, and Segregation Abstracts