Table of Contents  |  Search Technical Documentation  |  References

NAEP Sample Design → Sample Design for the 2000 Assessment → National Main Assessment Sample Design in 2000 → Student Sample Selection for the 2000 National Main Assessment → Initial Student Sample for the 2000 National Main Assessment

NAEP Technical DocumentationInitial Student Sample for the 2000 National Main Assessment

Every student whose name appeared on the student listing form (SLF) had an equal chance of being selected in the initial sample. Based on the number of students on the SLF, the student-sampling program first determined the student sample size for the initial sample and then systematically generated a sample of SLF line numbers of the required size. The sampling process was systematic in that every k^th line number was selected after a random start. The sampling interval used in the systematic sample was obtained by dividing the number of students on the listing form by the calculated sample size. The random start was a number between one and the sampling interval. The students selected for the initial sample were identified by the selected SLF line numbers.

The target student sample size within a school for the initial sample varied by school type and grade. The target sample size was 100 students for grades 4 and 12 public schools and 60 students for grade 8 public schools. Because private schools—which generally have small enrollment—were heavily oversampled, they were allowed to be selected multiple times in grade 8 (up to two times) and in grade 12 (up to three times). The use of multiple selections reduced the number of sampled schools needed to achieve the student sample. The target sample size for private schools was 60 students per selection. Thus, large private schools with three selections had 180 students in sample.

In general, the initial student sample consisted of all students if the number of students on the SLF was less than a maximum cutoff. Otherwise, the number of students in the initial sample was the target sample size. The maximum cutoff value was slightly larger than the target sample size to avoid samples that excluded only a handful of students in a grade. The maximum cutoff was

• 110 students for grades 4 and 12 public schools;

• 70 students for grade 8 public schools; and

• 62 students per selection for all private schools.

Printer-friendly Version