Table of Contents  |  Search Technical Documentation  |  References

NAEP Weighting Procedures → Weighting Procedures and Variance Estimation for the 2004 Assessment → Computation of Full-Sample Weights for the 2004 Assessment → School and Student Nonresponse Adjustments for the 2004 Assessment → Student Nonresponse Adjustment for the 2004 Assessment → Development of Final Student Nonresponse Cells for the 2004 Assessment

NAEP Technical DocumentationDevelopment of Final Student Nonresponse Cells for the 2004 Assessment

Similar to the school nonresponse adjustment, cell and adjustment factor constraints are in place to prevent unstable nonresponse adjustments or large adjustment factors. All initial weighting cells with either less than 20 participating students or adjustment factors greater than 2.0 were collapsed with suitable adjustment cells.

Initial weighting cells were generally collapsed in reverse order of the cell structure, meaning that collapsing was done starting with the variables which were the least important predictors of student nonresponse (the last in the nesting cell structure). For public schools, race/ethnicity cells were collapsed first within gender, school nonresponse cell, upper/lower grade classification, student disabilities or English language learner (SD/ELL) status/subject, and subject. If further collapsing was required, cells were next combined across gender, and school nonresponse cells. Cells for public schools were never collapsed across upper/lower grade classification, SD/ELL and subject status. Likewise, for private schools, race/ethnicity cells were collapsed first. If further collapsing was required, cells were then collapsed across gender, and school nonresponse cells. Cells for private schools were never collapsed across upper/lower grade classification, and subject.

Printer-friendly Version