Table of Contents  |  Search Technical Documentation  |  References

NAEP Sample Design → NAEP 2003 Sample Design → 2003 State Assessment Sample Design → Stratification of Schools in the 2003 State Assessment → Stratification Variables → Stratification by Minority Classification

NAEP Technical DocumentationStratification by Minority Classification

Minority classification was based on the two largest minority percentages (among Black, Hispanic, Asian, American Indian, and for the first time in 2003, non-Hispanic White students) within each jurisdiction-urbanization classification. Within each of these cells, the largest minority percentage and the second largest minority percentage, and the minority groups to which they belonged, were identified. Three different minority classification schemes were used, as follows:

Case 1: Urbanization classifications with less than 7 percent of students in the largest minority group were not stratified by minority enrollment (minority stratification value 0). The minority stratification variable is set equal to 0 for all schools within the urbanization cell.

Case 2: Urbanization classifications with greater than or equal to 7 percent of students in the largest minority, but with no more than one minority with greater than or equal to 15 percent, were stratified by ordering the sum of the percent minority enrollment for the largest and next largest minority within the urbanization classification and dividing the schools into three groups with about equal numbers of students per minority classification (minority stratification values 1, 2, and 3).

Case 3: In urbanization classifications with greater than 15 percent of students in both the largest and next largest minority group, minority classifications were formed with the objective of providing equal strata with emphasis on the two largest minority groups. The stratification was performed as follows. The largest minority group provided the primary stratification variable; the second largest minority group provided the secondary stratification variable. Within urbanization classification, the schools were first sorted based on the primary stratification variable. Then they were divided into two groups of schools containing approximately equal numbers of students based on estimated grade enrollment. Within each of these two groups, the schools were sorted by the secondary stratification variable and subdivided into two groups of schools containing approximately equal numbers of students. As a result, within urbanization classification there were four minority classifications (e.g., low primary variable/low secondary variable, low primary variable/high secondary variable, high primary variable/low secondary variable, and high primary variable/high secondary variable, levels 4, 5, 7, and 6, respectively). Note that the serpentine sort reverses the order by the second minority variable within the two cells generated by the first minority variable. That is why levels 7 and 6 are reversed.

Printer-friendly Version