To clarify the role of schools in reproducing social stratification, sociologists have focused mainly on high schools, especially curriculum tracking (see e.g., Gamoran & Mare, 1989; Gamoran, 1992). These practices have become less standard over the past two decades (Lucas, 1999), but they are still much in evidence in the United States (Kao & Thompson, 2003). Generally, tracking studies examine curriculum or course-taking differences within schools, comparing high- versus low-ability group placements (Gamoran, 1992; Hallinan, 1992; Stevenson, Schiller, & Schneider, 1994). Even so, family SES, which predicts tracks within schools, also predicts tracks between schools because the availability and quality of high-level programs vary from school to school, even in the same system (Spade, Columba, & Vanfossen, 1997), and family SES of students predicts this availability (Jones, Vanfossen, & Ensminger, 1995; Spade et al., 1997). Aside from comparisons of public and parochial schools, however, research on effects of between-school tracking is thin.

Ability-group tracking is found in a large majority of U.S. middle schools as well (Braddock, 1990), but research on tracking in middle schools is much less extensive than in high schools, (e.g., Cairns, Cairns, & Neckerman 1989; Eccles, Midgley, & Adler, 1984; Eccles et al., 1993; Feldman & Elliott, 1990; Reynolds, 1992; Simmons & Blyth, 1987). Still, it is in middle school that many students first experience formal tracking, a key organizational change for them (Braddock, Wu, & McPartland 1988; Hoffer, 1992, 1994).