Table of Contents  |  Search Technical Documentation  |  References

NAEP Assessment Weighting Procedures → Weighting Procedures for the 2007 Assessment → Computation of Full-Sample Weights for the 2007 Assessment → School and Student Nonresponse Adjustments for the 2007 Assessment → School Nonresponse Adjustment for the 2007 Assessment → Development of Final School Nonresponse Cells for the 2007 Assessment

NAEP Technical DocumentationDevelopment of Final School Nonresponse Cells for the 2007 Assessment

Limits were placed on the magnitude of cell sizes and adjustment factors to prevent unstable nonresponse adjustments and unacceptably large nonresponse factors. All initial weighting cells with fewer than six cooperating schools or adjustment factors greater than 3.0 for the full sample weight were collapsed with suitable adjacent cells. Simultaneously, all initial weighting cells for any replicate with fewer than four cooperating schools or adjustment factors greater than the maximum of 3.0 or 2 times the full sample nonresponse adjustment factor were collapsed with suitable adjacent cells. Initial weighting cells were generally collapsed in reverse order of the cell structure; that is, starting at the bottom of the nesting structure and working up towards the top level of the nesting structure.

For the state-based samples at grades 4 and 8, cells with the most similar racial/ethnic strata within a given jurisdiction/Trial Urban District Assessment (TUDA) district and urbanicity (locale) stratum were collapsed first. If further collapsing was required after all levels of racial/ethnic strata were collapsed, cells with the most similar urbanicity strata were combined next. Cells were never collapsed across jurisdiction or TUDA district.

For the national public school sample at grade 12, racial/ethnic status cells within a given census division stratum and urbanicity stratum were collapsed first. If further collapsing was necessary, cells with the most similar urbanicity strata were combined. Any further collapsing occurred across census division strata but never across census region.

Likewise for the private school samples, cells with the most similar racial/ethnic strata were collapsed first. If further collapsing was necessary, cells with the most similar urbanicity strata were combined. Any further collapsing occurred across census division, and, finally, across private school type.

Printer-friendly Version