English title: Comparison of Three Modes for a Crime Victimization Survey
Author(s): S. Laaksonen - Markku Heiskanen -
Type: Journal article
This paper compares three different modes of data collection: face-to-face, telephone, and web surveys for a national crime survey. The survey designs compared are as identical as possible. Our application area, crime victimization, is demanding in many ways, and not least because of the sensitivity of many of the questions. We first assess the effects that each mode has on frame coverage and on the nonresponse rate. We exploit this analysis in order to develop adjusted sample weights so that we can compare differences in measurement. Second, we hypothesize that the mode that better secures confidentiality will produce higher victimization prevalence. Our results suggest that sensitive questions were particularly difficult in telephone interviews. It seems that the representativeness of survey results depends on several factors—not solely, for instance, on the response rate.
From page no: 459
To page no: 483
Journal: Journal of Survey Statistics and Methodology