Contact Histories from the Survey of Grouped Individuals: Multivariate Multilevel Survival Analysis of Grouped Censored Data with Competing Risks

• Activity Number: 141
• Type: Contributed
• Date/Time: Monday, August 2, 2010 : 8:30 AM to 10:20 AM
• Sponsor: Section on Survey Research Methods
• Author(s): Hiroaki Minato*+
• Companies: NORC
• Keywords: censored data ; competing risks ; contact history ; dynamic modeling ; multilevel grouped data ; multivariate survival analysis

Abstract

Suppose we have a sample of households. Each household contains at least one person. A survey of our interest is to interview multiple randomly selected eligible persons in each randomly selected eligible household in the sample. Define that a household completes the survey if every selected person of the household responds to his or her questionnaire. Extending the current theory of multivariate survival analysis of grouped censored data, we propose to model the survey completion time of households and the survey response time of persons simultaneously. More specifically, we describe group response outcomes and times as well as individual response outcomes and times in terms of within-group dependency structures of static and dynamic types. Our goal is to construct a statistical framework useful for characterizing and understanding contact histories in survey research.