Working Paper: NBER ID: w12988
Authors: Jacob Vigdor; Jens Ludwig
Abstract: The mid-1980s witnessed breaks in two important trends related to race and schooling. School segregation, which had been declining, began a period of relative stasis. Black-white test score gaps, which had also been declining, also stagnated. The notion that these two phenomena may be related is also supported by basic cross-sectional evidence. We review existing literature on the relationship between neighborhood- and school-level segregation and the test score gap. Several recent studies point to a statistically significant causal relationship between school segregation and the test score gap, though in many cases the magnitude of the relationship is small in economic terms. Experimental studies, as well as methodologically convincing non-experimental studies, suggest that there is little if any causal role for neighborhood segregation operating through a mechanism other than school segregation.
Keywords: Segregation; Test Score Gap; Black-White Achievement Gap
JEL Codes: I20; J15; R23
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| school segregation (I24) | black-white test score gap (I24) |
| neighborhood segregation (R23) | black-white test score gap (I24) |
| neighborhood characteristics (R23) | educational outcomes (I26) |
| Moving to Opportunity (MTO) program (R28) | children's test scores (I21) |
| school racial composition (I24) | student achievement (I24) |
| segregation levels (Y40) | black-white test score gap (I24) |