NAEP Assessment Sample Design → NAEP 2009 Sample Design → Sample Design for the 2009 National Assessment → 2009 Twelfth-Grade Public School National Assessment

NAEP Technical Documentation2009 Twelfth-Grade Public School National Assessment

			Target Population Sampling Frame Stratification of Schools School Sample Selection Substitute Schools Ineligible Schools Student Sample Selection Participation

The sample design for the NAEP 2009 study in twelfth-grade public schools achieved a nationally representative sample of public school students in the twelfth grade through the use of a two-stage approach: selection of schools within strata and selection of students within schools. The first-stage sample of schools was selected with probability proportional to a measure of size based on the estimated grade-specific enrollment in the schools.

The general target for the twelfth-grade public school sample was 99,000 assessed students for the operational test in mathematics, reading, and science. This included two parts: an  11 state sample and a complement sample. The 11 state sample had a target sample size of approximately 58,000 students and was designed to give state-level estimates for reading and mathematics, as well as contributing to the national estimate for science. The complement sample covered the remaining 39 states, the District of Columbia, the Bureau of Indian Education (BIE) schools, and Department of Defense Education Activity (DoDEA) schools located within the 50 states and the District of Columbia. The target sample size for this component was approximately 41,000 students, and was designed so that when combined with the data from the eleven state-level samples, national estimates could be obtained for reading, mathematics, and science. The 11 states for which state level estimates were required for reading and mathematics are Arkansas, Connecticut, Florida, Idaho, Illinois, Iowa, Massachusetts, New Hampshire, New Jersey, South Dakota, and West Virginia. Target sample sizes were adjusted for expected school and student response and eligibility.

As in past assessments, Black and Hispanic students were oversampled at a moderate rate in the complement sample but not in the 11 states included in the state-level assessments. Schools in a high Black/Hispanic stratum (schools with 15 percent or more Black and Hispanic students) were sampled at twice the rate as those in a low Black/Hispanic stratum to implement oversampling of Black and Hispanics in the twelfth-grade public school sample.

From the stratified frame of twelfth-grade public schools, 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. Schools with a high enrollment of Black and Hispanic students were sampled at twice the rate by doubling the measure of size of these schools in the complement states. 

Each selected school in the twelfth-grade public school sample provided a list of eligible enrolled students from which a systematic, equal probability sample of students was drawn. The target sample size was 150 for the complement states and varied for state-assessment states (see student sample selection). If the school's twelfth-grade student total was only slightly larger than the target sample size (up to 120 percent of the target sample size), then all twelfth-grade students in the school were targeted for assessment.

Printer-friendly Version