Simulation Evaluation of Adaptive Survey Designs for a Community Health Survey

• Activity Number: 423 - Contributed Poster Presentations: Survey Research Methods Section
• Type: Contributed
• Date/Time: Tuesday, July 30, 2019 : 2:00 PM to 3:50 PM
• Sponsor: Survey Research Methods Section
• Author(s): David Brown* and F Jay Breidt
• Companies: Colorado State University and Colorado State University
• Keywords: Adaptive Survey Design; Auxiliary Information; Paradata; Responsive Survey Design; Response Propensity Models

Abstract

The Health District of Northern Larimer County conducts a triennial health survey of the adult population in Larimer County, Colorado. The survey questionnaire covers topics that range from basic health and nutrition to interactions with the health care and insurance systems. In its latest iterations (2013 and 2016), the survey has been conducted with an initial web push by mail and additional mail follow-up, with monetary incentives at some mail stages. Using data from previous iterations of the survey, demographic estimates from the US Census Bureau, and auxiliary information from local government agencies, we create a virtual adult population for the county. We use paradata from past surveys to construct response models that include estimated effects of timing of reminders and survey incentives on response rates of different groups of interest. We use the virtual population and its simulated response behavior to assess the impact on survey errors and survey costs of different adaptive survey designs, some of which might be adopted for experimental comparison in the actual survey implementation.