Table of Contents  |  Search Technical Documentation  |  References

NAEP Assessment Sample Design → NAEP 2007 Sample Design → Sample Design for the 2007 National Assessment → 2007 Private School National Assessment → Stratification of Schools for the 2007 National Private School Assessment

NAEP Technical DocumentationStratification of Schools for the 2007 National Private School Assessment

The stratification of the schools involved four dimensions. Explicit stratification for the NAEP 2007 private school sample was by private school type: Roman Catholic, non-Catholic, and Unknown Affiliation. Note that this division differs from previous years in which the non-Catholic stratum was further divided into specific school types. Private school affiliation was unknown for nonrespondents to the National Center for Education Statistics (NCES) Private School Universe Survey (PSS).

Within each explicit stratum, the private schools were hierarchically sorted by the following variables:

• Census region,
• urbanicity status,
• race/ethnicity status,
• and estimated grade enrollment.

The implicit stratification in this four-fold hierarchical stratification was achieved via a "serpentine sort."

Census region was used as the first level of implicit stratification for the NAEP 2007 private school sample.  The next level of stratification was an urbanization classification based on type of location, as defined by the PSS:

1. Large City: A principal city of a Metropolitan Core Based Statistical Area (CBSA), with the city having a population greater than or equal to 250,000.

2. Mid-size City: A principal city of a Metropolitan CBSA, with the city having a population less than 250,000.

3. Urban Fringe of a Large City: Any incorporated place, Census-designated place, or non-place territory within a Metropolitan CBSA of a Large City and defined as urban by the U.S. Census Bureau.

4. Urban Fringe of a Mid-size City: Any incorporated place, Census-designated place, or non-place territory within a CBSA of a Mid-size City and defined as urban by the Census Bureau.

5. Large Town: An incorporated place or Census-designated place with a population greater than or equal to 25,000 and located outside a Metropolitan CBSA or inside a Micropolitan CBSA.

6. Small Town: An incorporated place or Census-designated place with population less than 25,000 and greater than or equal to 2,500 and located outside a Metropolitan CBSA or inside a Micropolitan CBSA.

7. Rural, outside CBSA: Any incorporated place, Census-designated place, or non-place territory not within a Metropolitan CBSA or within a Micropolitan CBSA and defined as rural by the Census Bureau.

8. Rural, inside CBSA: Any incorporated place, Census-designated place, or non-place territory within a Metropolitan CBSA and defined as rural by the Census Bureau.

Urbanicity status cells within a geography stratum which did not meet the minimum of six expected hits were collapsed, as were the Census Division strata. The criterion for adequacy was that the cell had to correspond to a minimum of six expected hits. The urbanization variable was equal to the original type of location if no collapsing was necessary to cover an inadequate original cell. For example, type of location categories 1 and 2 are collapsed if either cell has fewer than six expected hits. That is, a single category would be created by combining Large City with Mid-size City. If these collapsed cells were still inadequate, they were further collapsed with the two types of Urban Fringe cells to form a single cell made up of large and mid-size cities and the urban fringes of large and mid-size cities. The urbanization variable was equal to the cell value of the final level of collapsing. Prior experience with this type of stratification has shown that the greatest efficiency of stratification results when cities and urban fringe areas are kept separate from towns and rural areas, even if the six expected hits condition is violated.

The last stage of stratification was a division of the geographic/urbanicity strata into race/ethnicity strata, if the expected number of hits was large enough (i.e., at least equal to 12). This was done by deciding first on the number of race/ethnicity strata and then dividing the geography/urbanicity stratum into this many race/ethnicity strata by using a sorted list based on percentage of students who are Black, Hispanic, and American Indian/Alaska Native and then divided into three sets using that ordering (lowest percentage, middle percentage, highest percentage), with roughly equal expected hits for each set.

Schools were sorted within stratification cells by estimated grade enrollment.

Printer-friendly Version