1352.0.55.113 - Research Paper: Empirical Models for Survey Cost, Response Rate and Bias Using Paradata (Methodology Advisory Committee), Nov 2010  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 02/03/2012  First Issue
   Page tools: Print Print Page Print all pages in this productPrint All
  • About this Release

The pressures of declining survey response rates and increasing operational costs pose a threat to the quality of the survey estimates. Various sources of paradata collected by interviewers, demographic information, survey data from previous collections, and administrative data give survey methodologists and practitioners unprecedented opportunity to understand respondent behaviour, and to analyse the effectiveness of survey operations, the associated cost structures, and the implications for survey estimate accuracy.
This paper presents modelling work for some ABS household and business surveys on response rates, survey cost structures, and potential bias resulting from changes to data collection inputs and efforts. Using these models, data collection operations can be re-configured by taking advantage of more cost effective follow-up methods and prioritising targets to improve response rates and survey estimate quality.