- Surveys & Programs
- Data & Tools
- Fast Facts
- News & Events
- Publications & Products
- About Us
Table of Contents | Search Technical Documentation | References
NAEP Technical DocumentationDevelopment of School Nonresponse Cells for Private School Samples
Nonresponse cells were established and nonresponse adjustment was done for the private schools that were assigned at least one “A” session. (See NAEP 2002 Sample Design for a description of sessions.) Preliminary weighting cells were formed for the private school sample based on the private school strata, with a modification as follows. The primary private school stratum was based on school affiliation (Roman Catholic, conservative Christian, Lutheran, and other private). The Private School Universe Survey (PSS) frame contained many schools for which the school affiliation was unknown, and these schools formed a separate set of strata. For purposes of school nonresponse, an affiliation was imputed for all of the unknown-affiliation strata schools,1 creating a revised private school stratum field in which all of the unknown-affiliation schools were assigned as Roman Catholic, Conservative Christian, Lutheran, or other private.
Weighting cells fulfilled the following criteria:
-
at least 6 sampled schools in the cell;
-
at least 4 sampled schools in the cell with nonzero replicate weights, for each replicate;
-
maximum school nonresponse adjustments and replicate school nonresponse adjustments of 4.0; and
-
a replicate school nonresponse adjustment no larger than 2 times the corresponding school nonresponse adjustment.
Preliminary cells that did not satisfy these criteria were combined with other cells.
1 There were a total of 38 of these schools. The NAEP sampling and weighting staff gathered the necessary information about school affiliation from the school personnel themselves (for 31 schools) or from personal knowledge about the school (for 7 schools).
Last updated 11 March 2009 (RF)
