As with the state operational samples, the general goal for the national twelfth-grade sample was to achieve a self-weighting sample at the student level. The target sample size in each school for the national sample component (the complement set of 39 states and the District of Columbia that did not participate in the state assessment program for twelfth grade) was 150.
The measure of size computation was similar to that used for the twelfth-grade state assessment, except that within the complement states, schools with a high proportion of Black/Hispanic students (those with greater than 15 percent Blacks and Hispanic students) had their relative measures of size doubled compared to schools with low numbers of Black/Hispanic students. For the twelfth-grade public school sample, a maximum of four "hits" were allowed per school.
Schools were ordered within each jurisdiction using the serpentine sort (by census division, urbanicity, race/ethnicity status, school type, and median income). A systematic sample was then drawn using this serpentine sorted list and the measures of size. A total of 576 schools was selected for the complement states from 18,217 eligible public schools. This sample of 576 schools was in addition to the 1,098 schools that were sampled from the 11 states in the state-assessment program: a total of 1,674 sampled twelfth-grade schools from the Common Core of Data (CCD) frame.
New and newly-eligible schools were sampled from the new-school frame. A total of 8 schools were selected from the 286 new schools identified in the complement stratum states. These 8 sampled new schools were in addition to 16 sampled new schools from the 11 states in the state assessment program: a total of 24 sampled new schools for the twelfth-grade public school national sample.
The overall twelfth-grade public school sample available for national estimates was 1,698 schools.