Skip Navigation

Table of Contents  |  Search Technical Documentation  |  References

NAEP Assessment Sample Design → NAEP 2006 Sample Design → Selection of Primary Sampling Units (PSUs) for the 2006 Assessment → Primary Sampling Unit (PSU) Generation: Non-Metropolitan Statistical Areas for the 2006 Assessment

Primary Sampling Unit (PSU) Generation: Non-Metropolitan Statistical Areas for the 2006 Assessment

A software algorithm was utilized to define a preliminary set of primary sampling units (PSUs) satisfying the constraints. The input set consisted of all of the non-metropolitan counties. The software formed PSUs that satisfied the minimum size constraints, respecting state boundaries (i.e., not crossing state boundaries). The software also minimized the maximum point-to-point distance for the candidate PSUs, while still satisfying the minimum size constraints (15,000 youths in the Northeast and South Census regions, and 10,000 youths in the Midwest and West Census regions). "Worst first" was the general approach: the county that had the PSU with the largest maximum point-to-point distance was fitted first, with those counties that best fit a PSU containing the "worst-first" county. The algorithm was then run on the remaining counties not yet assigned to a PSU.

Initially, there were 22 counties that could not be combined into PSUs that satisfied the minimum size constraints while still remaining within a single state. Nine of the PSUs (formed from 12 of these counties) that were below the minimum size requirement were allowed to stand, since satisfying the minimum size requirement was not reasonably possible. The remaining counties were in Alaska, for which PSUs were manually drawn to better respect interstate highways (being drawn along the axis of these highways) and mountain ranges (avoiding crossing of ranges with poor road access). For Alaska, the proto-PSUs created by the program were replaced by the PSUs created for the NAEP 2004 Assessment. The end result of this procedure was that all non-metropolitan PSUs were contained within state boundaries. There were a total of 670 final non-metropolitan PSUs.

The table below presents the number of PSUs, the number of counties represented, and the estimated number of youths (total and mean per PSU) by Census region. The estimated number of youths (persons age 0 to 17) for each county comes from the 2003 U.S. Census Bureau's Population Estimated Program.

Non-Metropolitan Statistical Area primary sampling unit (PSU) frame, by Census region: 2006
Census region PSUs Counties Youths Mean youths per PSU
Total 670 2,050 11,879,394 17,730
Northeast 50 94 1,176,274 23,525
Midwest 249 769 3,709,733 14,899
South 269 872 5,269,583 19,590
West 102 315 1,723,804 16,900
SOURCE: U.S. Department of Education, Institute of Education Sciences, National Center for Education Statistics, National Assessment of Educational Progress (NAEP), 2006.

Last updated 11 December 2009 (EH)

Printer-friendly Version