Borman, Geoffrey, & Dowling, Maritza
Schools and Inequality: A Multilevel Analysis of Coleman's Equality of Educational Opportunity Data
The University of Wisconsin
Reanalysis and reconceptualization of the Coleman report through a 2-level hierarchical linear model.
Journal Name or Institutional Affiliation:
Teacher College Record
Vol. 112, No. 5, pp. 1201-1246
- 40% of the variance in verbal achievement is attributable to differences across schools.
- There was a statistically significant positive relationship between stronger preferences among teachers to teach middle-class students and school mean achievement.
- Above and beyond the effects of other school-level resources, there are highly important contextual effects associated with attending more highly segregated schools with higher concentrations of poverty.
- The compositional effect of the school-level racial/ethnic context was nearly 1 1/3 times the magnitude of the student-level effect of being African American.
- Schools do indeed matter, in that when one examines outcomes across the national sample of schools, fully 40% of the variability in verbal achievement lies between schools.
- A large proportion of the variation among true school means is related to differences that are explained by school characteristics.
- Substantial school-to-school variability in terms of the within-school social distributions of achievement.
- Both the racial/ethnic and social class composition of a student's school are approximately 150% more important than a student's individual race/ethnicity or social class for understanding educational outcomes.
Academic Achievement, Classroom Composition, Curriculum, Peer Effects, Racial Composition, SES Composition, Teachers
Secondary Survey Data
Method of Analysis:
Unit of Analysis:
- Coleman's EEO data files maintained by the Inter-university Consortium for Political and Social Research (ICPSR).
- Sample of over 570,000 students from grades 1,3,6, 9 and 12.
- Survey responses from about 4,000 principals and 40,000 teachers.
- Final sample 30,590 students and 226 schools
- IV: Student Level Variables, School-level variables, Science laboratory facilities, extracurricular activities, comprehensiveness of the curriculum, volumes per student, movement between tracks, etc.
- DV: Student's score on a standardized verbal ability test.