In state assessments (mathematics, reading, science, and writing), a sample of schools and students is selected to represent each participating state. In an average state, 2,500 students in approximately 100 public schools are assessed per grade, for each subject assessed. The selection process for schools uses stratified random sampling within categories of schools with similar characteristics. Some schools or groups of schools (districts) may be selected for each assessment cycle if they are unique in the state. For instance, if a particular district is in the only major metropolitan area of a state or has the majority of a minority population in the state, it may be selected for assessment more often. Additionally, even if a state decides not to participate at the state level, schools in that state identified for the national sample will still be asked to participate.
For the national assessments, probability samples of schools and students are selected to represent the diverse student population in the United States. The numbers of schools and students vary from cycle to cycle, depending on the number of subjects and items to be assessed. A national sample will have sufficient schools and students to yield data for public schools, each of the four NAEP regions of the country, as well as sex, race, degree of urbanization of school location, parent education, and participation in the National School Lunch Program (NSLP). A separate grade 12 sample of schools is also selected to produce national and regional estimates, as state NAEP does not yet include grade 12 (a pilot study of grade 12 state NAEP will be held in 2009). A national sample of nonpublic (private) schools is also selected for grades 4, 8, and 12. This sample is designed to produce national and regional estimates of student performance for private schools.
Typically, 30 students per grade per subject are selected randomly in each school. Some of the students who are randomly selected may be classified as students with disabilities (SD) or as English language learners (ELL). NAEP's goal is to assess all students in the sample. For a description of accommodation issues, read about NAEP's inclusion policy.
To find out more about NAEP sampling methods, see details about sampling for the reading assessments, or explore the frequently asked questions about the state sample design.
Note that beginning with the 2002 NAEP assessments, a combined sample of public schools was selected for both state and national NAEP. From the schools representing 50 states, a sub-sample is identified as the national subset. Drawing a subset of schools from all of the state samples to produce national estimates reduces the burden of testing by decreasing the total number of schools participating in state and national NAEP.
This sampling information applies only to the main NAEP assessments. The long-term trend assessments, scheduled every four years, use a nationally representative sample and do not report results by state.