The 2005 NAEP state operational mathematics, science, and reading assessments covered fourth- and eighth-grade students in public schools for each subject, referred to as the alpha sample. A representative sample of public school students was drawn in each participating state, the District of Columbia, Puerto Rico, Bureau of Indian Affairs (BIA) schools, Department of Defense (DoD) schools and in school districts chosen for the Trial Urban District Assessment (TUDA) study. State participation in the mathematics and reading assessments was mandatory, whereas participation in the science study was optional, with eight states declining to participate. By design, Puerto Rico participated only in the mathematics state assessment (not in the reading or science state assessments.)
Each jurisdiction sample was designed to produce aggregate estimates with approximately equal precision for all the participating jurisdictions, as well as estimates for various subpopulations of interest. The target student sample size per jurisdiction for an operational assessment was 3,150 students per grade. Thus, jurisdictions which chose to participate in all three assessment studies had a target sample of 9,450 students per grade, whereas those which excluded the science assessment had a sample of 6,300 students per grade.
The target population for the 2005 state assessment program included students in public schools who were enrolled in the fourth and eighth grades at the time of assessment. The sampling frame included public schools having the relevant grade in each jurisdiction. The samples were selected based on a two-stage sample design: selection of schools within participating jurisdictions and selection of students within schools.
From the stratified frame of public schools for each grade within each jurisdiction, a systematic random sample of grade-eligible schools was drawn with probability proportional to a measure of size based on the estimated grade-specific enrollment of the school.
For the TUDA study, additional schools were sampled from the 10 participating TUDA districts at the same time schools were selected for the jurisdiction samples. The TUDA districts were the nine districts which participated in 2003 (Los Angeles, CA; San Diego, CA; Atlanta, GA; Chicago, IL; Boston, MA; Charlotte, NC; New York City, NY; Cleveland, OH; Houston, TX), along with a tenth district in 2005: Austin, TX. These subsamples affected the design of the state samples in those states where TUDA districts were oversampled. In each of these states, there were distinct sampling rates for each TUDA district and for the rest of the state.
Each selected school provided a list of eligible enrolled students from which a systematic sample of students was drawn. In states that chose not to participate in the science assessment, 60 students, if possible, were selected from each school (30 students for mathematics and 30 for reading). In states which chose to participate in the science assessment, 90 students (if possible) were selected from schools (30 students for each of the subject areas). In smaller schools, all students were sampled, and divided equally among the relevant subjects.