NAEP Sample Design → Sample Design for the 2000 Assessment → National Main Assessment Sample Design in 2000 → Sampling of Schools in the 2000 National Main Assessment → Assigning Measure of Size and Selecting School Samples for the 2000 National Main Assessment Assigning Measure of Size and Selecting School Samples for the 2000 National Main Assessment
To increase cost efficiency in sampling, the 2000 national main samples were designed to include schools with a relatively high number of students and more public schools with higher percentages of minority students. A measure of size was assigned to each school based on two factors—school type (public and private) and grade:
In the table above, Si is the estimated number of grade-eligible students in a given school i; L is the target within-school student sample size for a given grade and school type; and ki is the ith school's oversampling factor; where
This procedure was used to obtain approximately self-weighting samples of students (i.e., students selected with approximately equal overall probabilities) at each grade and school type within oversampling domains. Two aspects of the school-level measure-of-size cause the resulting student weights to vary:
For each grade and public/private school sample, schools were selected (without replacement) within the appropriate sample areas systematically from a sorted list with probabilities proportional to assigned measures of size. Schools were ordered to achieve an implicit stratification on school characteristics related to school achievement and to ensure that the sampled schools represented a variety of population student groups. To meet specific sample requirements, the school sampling procedures differed by
Last updated 19 March 2008 (GF) |