Assessment of Total Survey Error for the National Immunization Survey (NIS) 2009--2010

• Activity Number: 88
• Type: Contributed
• Date/Time: Sunday, July 29, 2012 : 4:00 PM to 5:50 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Vicki Jean Pineau*+ and Kirk M. Wolter and Benjamin Skalland and Wei Zeng and Meena Khare and Zhen Zhao
• Companies: NORC and NORC and NORC and NORC and National Center for Health Statistics and National Center for Immunization and Respiratory Diseases
• Keywords: Noncoverage ; Nonresponse ; Monte Carlo Simulation

Abstract

Total survey error (TSE) is the sum of errors that arise at every step of a survey. It includes both sampling and non-sampling errors (e.g., population coverage, nonresponse, and measurement error). In this study, we assess TSE in estimated vaccination coverage rates derived from the NIS, which is conducted annually by the Centers for Disease Control and Prevention. We utilize 2009-2010 NIS data for children 19-35 months and adolescents 13-17 years, specify a distribution function for each component of sampling and non-sampling error, and derive estimates of these component distributions from the best sources available such as the American Community Survey and the National Health Interview Survey. We apply a Monte Carlo simulation-based approach to combine the components of error into a total error distribution for each estimate examined. Findings suggest small biases exist (-0.2 to 2.1 percentage points) in the 2010 NIS vaccination coverage rate estimates for children in spite of escalating cell-phone-only households in the U.S. Bias in vaccination coverage rate estimates for adolescents was larger than that for children (over-estimation of 1.3 to 6.7 percentage points).