Coverage analysis in mix mode designed surveys

Summary: Statistics Netherlands is currently re-designing its household surveys.
All household surveys will gradually be converted from their singlemode
designs to a general mixed-mode design. In a survey with a mixed-mode
design we face mode effects of a varying nature. To produce high quality statistics,
these mode effects need to be taken into account in the design and the
estimation stage. In this paper we look at one of the sources underlying these
mode effects, namely the coverage errors.