IBUC 2006 Abstract

The Selection of Strata in Nonresponse Adjustment
Barry Schouten ; Statistics Netherlands, The Netherlands

Nonresponse in household surveys can be a threat to the quality of statistics. Research shows that often the
response to these surveys is selective with respect to demographic characteristics like age and household
composition. For this reason estimators are usually adjusted to account for nonresponse. The Bascula tool is
added to Blaise in order to support nonresponse adjustment.

Nonresponse adjustment methods make use of covariates that are available for both respondents and nonrespondents.
A problem is the selection of covariates that relate both to the key survey questions and to the
response behaviour. Therefore, often the process of selection is performed in two steps. First, candidate covariates
are selected that relate either to the nonresponse mechanism or to the key survey topics. Second, adjustment is
performed using these candidate covariates using for instance Bascula.

We present a classification tree method that allows for the construction of weighting strata that simultaneously
account for the relation between response behaviour, survey questions and covariates.
The method is computationally intensive but it facilitates nonresponse adjustment in a single step. The method may
be included in Blaise to combine selection of strata and weighting in one tool.

Contact: bstn@cbs.nl