Might a Social Integration Navigator Guide Response Propensity Models in Nonresponse Follow-Up?

• Activity Number: 632
• Type: Topic Contributed
• Date/Time: Thursday, August 8, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Survey Research Methods Section
• Author(s): Asaph Young Chun and Kevin Zajac*+
• Companies: US Census Bureau and US Census Bureau
• Keywords: nonresponse ; response propensity ; paradata ; mode ; census

Abstract

The purpose of this paper is to identify paradata and multi-dimensional factors affecting nonresponse in the U.S. decennial census by using social integration (or social isolation) as a theoretical navigator. A person's response propensity has been empirically found to reflect the extent to which one is socially integrated (Abraham, Maitland and Bianchi, 2005) or isolated (Groves and Couper, 1998; Chun, 2009). We develop response propensity models specific to the Census nonresponse followup universe by tailoring to: 1) a followup mode of telephone, in-person, and administrative records, and 2) nonresponse outcomes such as occupied nonresponse household, vacancy and deletes, as well as proxy respondent. We apply the concept of "pandata," the data linked among multiple quality data sources which include paradata, census and administrative records, to craft robust response propensity models. We will use response data files from the 2010 Census, and then merge paradata associated with nonresponse followup (e.g. contact history) as well as household and neighborhood characteristic paradata borrowed from the 2010 Planning Database (e.g., household structure, poverty measures).