Development and Testing an Interactive Voice Response Questionnaire for the NCVS

• In Event: Methodological Research to Support the National Crime Victimization Survey, 2008-2013

Abstract

The purpose of this project was to assess the feasibility of using Interactive Voice Response (IVR) to administer the National Crime Victimization Survey (NCVS). The project considered two different applications – one for the national survey and one for local surveys. Three research questions were addressed: 1) what are the response rates for the IVR?; 2) can the IVR be adapted for use on the NCVS; and 3) what are the victimization rates for the IVR. To address these questions, inbound and outbound IVR versions of the NCVS were compared to a telephone interview in two metropolitan areas. The results suggest that using IVR to administer the NCVS victimization screener (NCVS-1) is feasible and should be considered as a way to follow-up respondents for the ongoing NCVS or for local area surveys. The response rates for this portion of the interview were comparable to a telephone interview and, with some exceptions, IVR was successfully adapted to the NCVS. The number of persons reporting an incident on NCVS-1 was the same as the telephone interview. Using IVR to administer the NCVS-2 posed more challenges. Approximately 30% of respondents did not complete all forms and particular items posed measurement problems.

Authors

• David Cantor, Westat

• Douglas Williams, Westat