In order to address both these goals for the second study year, a second round of random assignment was conducted consisting of two groups of students, applicants and nonapplicants (2stage in fall 2006, see Stage 2 of Figure ES.1). The application process in the sec-ond year of the study was conducted the same as in the first year of the study. Applicants in the second year consisted of newly identified low-performing students who were new appli-cants in year 2 and students from Cohort 1 who voluntarily applied to the second year of the study. Both of these groups of student applicants in Year 2 were randomly assigned by grade within each after-school center to either the enhanced program group or the regular program group; applicants from Cohort 1 were also randomly assigned by their first year treatment status (whether they were part of the enhanced or regular after-school program group). Ran-domly assigning for a second time students who participated in the first year, rather than allowing them to maintain their initial randomly assigned grouping, ensured that those who were offered the enhanced program the first year did not receive special treatment once the study was extended.

Nonapplicants are the remaining Cohort 1 students who had participated in the first year of the study, but did not apply to the second year of the study. They too were randomly as-signed (separately from applicants) by grade and their first year treatment status within each after-school center. Randomly assigning both the applicants and nonapplicants from Cohort 1 maintains an intent-to-treat sample of Cohort 1 students who are cumulatively offered two years of the program or never offered the program. (Note, fifth-graders from Cohort 1 were excluded from the second stage of the random assignment in fall 2006 because, as sixth-graders, they were no longer eligible for the program and thus did not reapply.)