249 – Estimation and Inference Methods with Complex Survey Data
When Do "Do it Yourself-ers" Really Do It Themselves? Using Paradata to Explore the Preference for Self-completion Modes in a Multi-mode Survey
Sara Zuckerbraun
RTI International
Melisssa Hobbs
RTI International
Angela Greene
RTI International
Lauren Harris-Kojetin
National Center for Health Statistics
Manisha Sengupta
National Center for Health Statistics
The NSLTCP includes a mixed mode (mail, web, telephone) survey of approximately 17,000 residential care facilities and adult day services centers in the U.S. The survey is planned to be conducted biennially, and its analytic goal is to collect information from these providers which is unavailable from other sources. We hypothesized that, of the modes available to complete the survey, respondents would prefer self-completion modes over telephone. In this analysis we used paradata from call records to gain empirical evidence to test our hypothesis. As other projects have done, we will use the results of this analysis to optimize our data collection strategy for future years of the survey (Kreuter et.al.,2010). We first examined mode completion choices and assessed whether they varied by provider size or type (adult day versus residential care). We then focused on the effectiveness of protocols that telephone interviewers followed that were intended to bring two groups of respondents to completion: 1)respondents who started by web but broke off; 2) cases that did not start but when contacted by telephone interviewers stated their intention to self-complete by web or paper ("Do-it Yourself-ers").