Gamoran, Adam, & An, Brian
Effects of School Segregation and School Resources in a Changing Policy Context
University of Wisconsin - Madison
Impact of school composition and school resources on student achievement growth.
Journal Name or Institutional Affiliation:
Presentation at meeting of the International Sociological Association Research Committee on Social Stratification and Mobility, Los Angeles, August
- Students in schools with high proportions of poor Blacks are negatively selected on achievement, but the causal relation is neutral or positive, not negative. Enhanced option schools, which bring extraordinary resources to the challenges of teaching students in high-poverty, racially isolated schools, do add additional value to mathematics achievement scores as compared with attendance zone schools. However, the added value is small compared to the size of the persisting achievement gap.
- With respect to causal effects, find positive consequences of entering racially isolated but not high -poverty schools, and this advantage, importantly, is not explained by enhanced option schools, which exhibit a positive but non-significant coefficient.
- Find that the combination of racial isolation, poverty concentration, and enhanced option schools add more to test scores than a racially mixed school without compensatory resources. Thus, results suggest that resources may be a more potent strategy than racial mixing for reducing achievement inequality. But the amount of value is quite small compared to the size of the gap.
- The authors found that achievement scores for all students increased but achievement gaps remained relatively the same with a slight reduction in inequality in reading. They concluded that resources do help reduce the achievement gap, the value added was small compared to the size of the gap.
Achievement Gap, Desegregation, Funding, Magnet Schools, Math, No Child Left Behind, Racial Composition, Reading
Secondary Survey Data
Method of Analysis:
Nashville school district middle school students
Unit of Analysis:
- Sample includes around 5,000 students in each grade level (3-8) over the five school years from 1998-99 to 2002-02.
- Close to 30,000 students between grades 3-8 in Nashville school district for 5 years from 1998-99 to 2002-03.
- Data includes 5,843 students with five test scores, 8,907 with four, 10,955 with three, and 14, 661 with two.
- A total of 110 schools are represented in the sample. The schools included 5 enhanced option schools and 16 magnet schools, the remainder were attendance-zone schools for all or part of the time period.
- Student background data drawn from school district files.
- DV: Data achievement based on scores of Tennessee Comprehensive Assessment Program (TCAP) - standardized tests of reading and mathematics for all students in grade 3-8. Test scores are available even when students change schools.
- Separate models for math and reading achievement scores.
- IV: individual background characteristics (race/ethnicity, sex, free lunch status, English language learners, disability status, suspensions/expulsions, and whether the student changed schools during the school year), school type (enhanced option and magnet versus zoned), and a school racial and economic composition (proportion Black and proportion of free or reduced-price lunch).
- Estimate multilevel fixed-effects models in which test scores are nested within students. The model incorporates the characteristics of each student's school separately in each year.
- Examines the effects of segregation versus resource allocation. Some argue that resources used in busing and desegregation could be better used to provide extra resources at school for disadvantaged populations. Nashville, TN ended desegregation and used enhanced option schools that provided extra resources to high-poverty, racially isolated schools.