Noncoverage Adjustments in a Single-Frame Cell-Phone Survey: Weighting Approach to Adjust for Phoneless and Landline-Only Households

• Activity Number: 41
• Type: Contributed
• Date/Time: Sunday, August 3, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Survey Research Methods Section
• Author(s): Nadarajasundaram Ganesh*+ and Meena Khare and Elizabeth Ormson and Wei Zeng and Jenny Jeyarajah and David Yankey and Zhen Zhao and Kirk Wolter
• Companies: NORC at the University of Chicago and NCHS/CDC/DHHS and NORC at the University of Chicago and NORC at the University of Chicago and CDC and CDC and CDC and NORC at the University of Chicago
• Keywords: cell-phone sample ; dual-frame sample design ; noncoverage ; RDD telephone surveys ; single-frame sample design ; weighting

Abstract

In 2011, the National Immunization Survey (NIS) began using a dual-frame landline and cell-phone sample design to monitor vaccination coverage rates among children 19-35 months. As of 2012, coverage of the cell-phone sampling frame has increased to 92.9% of all age-eligible children in the NIS. Thus, a single-frame cell-phone sample design could be a viable option for the NIS, but such a sample design would not cover children living in landline-only and phoneless households. In order to adjust for the noncoverage of children living in landline-only and phoneless households, we propose two distinct adjustment methods. The first option is to identify a subset of cell-phone households with sampled children who have similar socio-demographic characteristics as those from landline-only and phoneless households and use a weighting class based approach to adjust for the noncoverage. The second option is to adjust for the noncoverage by using socio-demographic characteristics in a raking adjustment step. We implemented the proposed methods with 2012 NIS data and compared survey estimates generated under a single-frame cell-phone sample design with similar estimates from the current dual-frame sample design.