USA Sample Design

The mental health stigma items will be a module on the 2006 General Social Survey. The GSS is a full-probability, in-person interview of adults living in households in the United States . The target sample size for these items is 1500. The field period should run from the start of February until the end of May. Data should be ready for distribution by December, 2006.

NORC uses a multi-stage, area probability sample. First, primary sampling units (PSUs) are selected. They are metro areas or non-metro counties. Next, segments are selected within PSUs. They are roughly equivalent to neighborhoods. There are at least 3 segments per PSU and in the larger, self-representing PSUs (e.g. Chicago or LA) there are more segments proportional to size. Then all housing units in a segment are listed and a random number of them will be drawn for the 2006 GSS. These households are then contacted and all household members are listed in a household enumeration form. Those members 18+ are then relisted in a Kish table and a respondent is randomly selected.

The GSS design is self-weighting for households, but to generalize to adults a weight based on the inverse of the number of adults in the household is needed. No post-stratification weights are used as the GSS's coverage of most demographic groups is close to Census targets.

For more details on sampling and weighting, see Appendix A of the GSS Cumulative Codebook. An on-line version can be found at www.icpsr.umich.edu

1022 E. Third Street, Bloomington, IN 47405 (812) 855-3841

Last updated: 6 September 2005
Comments: sgcmhs@indiana.edu
Copyright 2004, The Trustees of Indiana University
Copyright Complaints