Select the Participants

NAEP assesses representative samples of students rather than the entire student population. The sample selection process utilizes a probability sample design. In this type of sample, each school and each student has a known probability of being selected. Samples are selected according to a multistage design, with students drawn from within sampled public and private schools nationwide. The school probabilities are proportional to the estimated number of students in the grade assessed.

1. Identify Schools: Schools are identified based on the U.S. Department of Education's public school system database. Because national-level data on schools can be up to three years old, school data are verified with state departments of education. A national sample of nonpublic (private) schools is also selected for grades 4, 8, and 12.
   
2. Classify Schools: Schools are classified into groups by type of location (city, suburb, town, or rural) and then by racial/ethnic composition. These groups are called strata. 
   
3. Arrange Schools: Schools are sorted by a student achievement measure (e.g., school-level results on state achievement tests) to ensure NAEP represents all levels of school performance.
   
4. List Schools: The school groupings are then placed into a comprehensive ordered list. The specific schools that comprise the sample are ordered based on location type, race/ethnicity, and student achievement.
   
5. Select Schools: NCES draws a separate sample of schools from each stratum with probability proportional to school size. Small schools, high minority schools, and private schools are sampled to ensure that they are adequately represented. If the school participation rate in the sample is below 85 percent, the results cannot be reported.
   
6. Confirm Schools: A list of selected schools is sent to each state department of education to confirm eligibility. School closure or no eligible students would prevent a school from being selected.
   

Within each selected school and grade to be assessed, students from the selected grade (4, 8, or 12) are chosen at random to participate in NAEP for up to two subjects. Every student has a chance of being chosen--regardless of race/ethnicity, socioeconomic status, disability, status as an English language learner, or any other factors.

Results are never reported for individual students. National results are reported for all assessment subjects. Mathematics, reading (and sometimes science and writing) results are also reported at the state level and for participating urban districts. Both public and private schools results are reported for the nation, but only public school results are reported at the state level.

Students in a given assessment session will take a variety of different assessment forms, generally across several subjects. Individual students only complete a subset of NAEP items within each subject area.

A technique called Balanced Incomplete Block (BIB) spiraling is used in which groups of items are systematically arranged with other groups of items within test forms to assure that the entire content domain for the subject is covered. And all items are completed by a representative sample of students, though individual students only complete a fraction of the items.