Table of Contents  |  Search Technical Documentation  |  References

NAEP Weighting Procedures → 2002 Weighting Procedures and Variance Estimation → State Assessment Weights → State NAEP School and Student Nonresponse Adjustments → Computation of School Nonresponse Adjustments

NAEP Technical DocumentationComputation of School Nonresponse Adjustments

			Development of School Nonresponse Cells

From our school response experience in the 2002 state assessment, response rates varied widely across jurisdictions. Adjustment for school nonresponse was accomplished through a weighting cell adjustment.

The following nesting cell structure was used to define nonresponse weighting cells for each participating jurisdiction:

• Trial Urban District Assessment (TUDA) district versus all other districts for states with TUDA districts;
• urbanicity stratum; and
• minority status stratum.

When nonresponse cells based on this nesting structure were too small to allow for stable nonresponse adjustment, cells were collapsed. The final school nonresponse weighting cells within each jurisdiction j are designated as SNRADJ_c, c = 1,...,C, where C is the total number of weighting cells designated for the jurisdiction. The school nonresponse adjustments were computed as follows for each nonresponse weighting cell:

where

• S_c is the set of all eligible original schools in cell c;

• R_c is the set of all responding schools within S_c;

•   is the inverse of the school probability of selection;

• TRIM_s is a trimming factor for new schools;

• x_s is the school's estimated enrollment; and

• SUBADJ_s is a factor for differing enrollment between substitute schools and the originally sampled schools (see School Base Weights for Substitute Schools).¹

¹ Note that TRIM_s and SUBADJ_s are 1 for schools which are not trimmed and are not responding substitutes, respectively.

Printer-friendly Version