Assessment of Total Survey Error for the 2011 National Immunization Survey

• Activity Number: 695
• Type: Contributed
• Date/Time: Thursday, August 8, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Survey Research Methods Section
• Author(s): Wei Zeng*+ and Kirk Wolter and Benjamin Skalland and Vicki Pineau and Carla Black and Christina Dorell and Meena Khare
• Companies: NORC at the University of Chicago and NORC at the University of Chicago and NORC at the University of Chicago and NORC at the University of Chicago and Centers for Disease Control and Prevention and Centers for Disease Control and Prevention and NCHS/CDC
• Keywords: Monte Carlo ; Dual-Frame ; RDD

Abstract

Total Survey Error (TSE) is the sum of sampling error and the nonsampling errors that arise at every step of a sample survey. Since 2006, TSE modeling has provided a framework for evaluation of potential biases in the National Immunization Survey (NIS), a project conducted on an ongoing basis by the Centers for Disease Control and Prevention to measure the vaccination status of young children and adolescents. For the 2011 TSE analysis, we utilize revised inputs from the National Health Interview Survey (NHIS) that account for differing provider reporting methods between the NIS and the NHIS. Also for 2011, the NIS was fielded as a dual-frame RDD landline and cellular telephone sample, marking the first year TSE is assessed under the new dual-frame design. For TSE modeling, we specify a distribution function for each component of error, derive estimates of the component distributions from the best sources, and then apply a Monte Carlo simulation-based approach to combine the components of error into a total error distribution for each estimate examined. We compare estimates of bias derived from the TSE analysis for 2011 to bias obtained from TSE analyses conducted in 2009-2010.