To select sample schools, the surveyors first chose schoolswith twelfth grades. Then, for each school containing a twelfth grade, they identified the middle and elementary schools that “fed” their students into the secondary school. If a lower-grade school fed more than 90% of its students into the selected twelfth-grade school, then it was sampled with certainty; other lower-grade schools were sampled in proportion to the share of their students who were fed into the twelfth-grade school. The Coleman data con- tain a school identifier variable unique to each sampled school con- taining a twelfth grade. For students in lower-grade schools, this identifier refers to the sampled twelfth-grade school intowhich the students.
