Table of Contents  |  Search Technical Documentation  |  References

NAEP Sample Design → NAEP 2002 Sample Design → 2002 State Assessment Sample Design → School Sample Selection for the 2002 State Assessment → New-School Selection → New-School Samples

NAEP Technical DocumentationNew-School Samples

Any jurisdictions which were take-all-students or take-all-schools for the main frame sample were designated to be the same for the new-school sample.

For the remaining new schools in non take-all-school jurisdictions, we computed measures of size (for school s in jurisdiction j):

where  is the probability of selection of the school’s district into the new-school district sample, and E_js is the measure of size computed based on enrollment. If z_js was less than 1, then it served as the new school’s probability of selection. If z_js was greater than 1, then in some cases we trimmed it.¹

The new-school samples were implicitly stratified using a hierarchical ordering by jurisdiction, district, and estimated grade enrollment.

¹For fourth grade, all jurisdiction z_js values were trimmed to 3 if they were greater than 3, except for one jurisdiction which had a trimming point of 8. In eighth grade, the trimming level was generally 3, except for four jurisdictions which were higher (up to a maximum of 8).

Printer-friendly Version