Combining landline and mobile phone samples A dual frame approach
Abstract
More and more households abandon their landline phones and rely solely on cell phones. This implies a
challenge for survey researchers: since the cell phone only households are not included in the frames for
landline telephone surveys, samples based on these frames are in danger to be seriously biased due to
undercoverage, if respondents who do not have a landline are systematically different from respondents
who have a landline. Thus, strategies for combining samples from different frames need to be developed.
In this paper we give theoretical foundations for a dual frame approach to sampling, explain how samples
can be optimally allocated from these two frames, and describe an empirical application of a survey
conducted in Germany that used a dual frame approach.