Using different comparison groups than in earlier work, we find that ability tracking in math classes has little effect on growth in student math achievement. Previous articles use dummy variables to compare students from classes of high, middle, and low ability to the average heterogeneously grouped student. This paper furthers the research by controlling for ability groups at each type of school and establishes that tracking schools differ little from non-tracking schools in terms of student achievement growth. The paper also shows that the allocation of students and resources among classes is remarkably similar between the two types of schools.

---

---

---