Working Paper: NBER ID: w30370
Authors: Kate Antonovics; Sandra E. Black; Julie Berry Cullen; Akiva Yonah Meiselman
Abstract: Little is known about the pervasiveness or determinants of within-school ability tracking in the US. To fill this gap, we use detailed administrative data to estimate the extent of tracking in Texas public schools for grades 4 through 8 over the years 2011-2019. Strikingly, we find that ability tracking across classes within schools overwhelms sorting by ability across districts and schools, as well as sorting by race/ethnicity or economic disadvantage. We also examine how schools operationalize tracking as well as the local characteristics that predict tracking. Finally, we explore how exposure to tracking (and the bundle of associated practices) relates to achievement gains, finding that, on average, tracking increases inequality by slightly improving test scores of higher-achieving students without harming those of lower-achieving students.
Keywords: No keywords provided
JEL Codes: H75; I21; I24; I28
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Tracking by ability within schools (I24) | Variation in prior scores explained by classes (C38) |
| Tracking (L87) | Test scores for higher-achieving students (D29) |
| Tracking (L87) | Test scores for lower-achieving students (I24) |
| Exposure to middle school tracking (I24) | Predicted test scores five years after third grade for students initially at the 75th percentile (I21) |
| Tracking reduces the likelihood of initially low-achieving students being placed in advanced math courses (I24) | Tracking increases the likelihood for high-achieving students (D29) |
| Students exposed to more tracking (I24) | Greater inequality in peer achievement across classes (I24) |
| Students exposed to more tracking (I24) | Smaller average class sizes (I21) |