NAEP Technical DocumentationStratification of Public Schools in the 2008 Long-Term Trend (LTT) Assessment

Separate implicit stratification schemes were used to sort schools in certainty primary sampling units (PSUs) and noncertainty PSUs. In all cases, the implicit stratification was achieved via a "serpentine sort."

For certainty PSUs, the schools were hierarchically sorted by

• Census region,
• urbanization classification,
• race/ethnicity strata (less than 15 percent Balck and Hispanic students, 15 percent or less Black and Hispanic students), and
• estimated number of age eligible students.

If there were fewer than two expected sampled schools for a particular urbanization cell (in other words, the aggregated probabilities of selection over the schools in the cell on the frame is less than or equal to two), the cell was collapsed with a neighboring urbanization cell. If the expected sampled schools equaled or exceeded four, the race/ethnicity stratum was defined based on the summation of school percentages of Black, Hispanic, and American Indian/Alaska Native students. The race/ethnicity stratum was defined so that there were at least two expected sampled schools for each urbanization stratum (nested within the Census region). If the type of location stratum had an expected sample size less than four, no race/ethnicity strata were generated, and the final sort variable was the summation of school percentages of Black, Hispanic, and American Indian/Alaska Native students rather than estimated age eligible students.

Schools in noncertainty PSUs were hierarchically sorted by

• PSU stratum,
• urbanization classification,
• percent race/ethnicity. 

Race/ethnicity percentages for each school are the summation of the school percentages of Black, Hispanic, and American Indian/Alaska Native students.

Printer-friendly Version