![[ Visit Client Website ]](images/banner.gif)
Tuesday, March 31, 2009
Grand Hall area
Background:
Random-digit-dial (RDD) telephone surveys are experiencing both 1) declining response rates and 2) increasing under-coverage due to the prevalence of households that substitute a wireless telephone for their residential landline telephone. These changes increase the potential for bias in survey estimators and heighten the need for survey researchers to evaluate the sources and magnitudes of potential bias.
Objectives:
We estimate the distribution of total survey error in the National Immunization Survey (NIS) using special studies of components of error and Monte Carlo simulation.
Methods:
We apply a Monte Carlo simulation-based approach to assess bias in the NIS, a land-line telephone survey of 19-35 month-old children used to obtain national vaccination coverage estimates. We develop a model describing the survey stages at which component nonsampling error may be introduced due to nonresponse and under-coverage. We use that model and components of error estimated in special studies to quantify the extent to which noncoverage and nonresponse may bias the vaccination coverage estimates obtained from the NIS and present a distribution of the total survey error.
Results:
Results indicated that total error is distributed with a mean of 1.72 percent (95% CI: 1.71%, 1.74%) and final adjusted survey weights corrected for this error. While small, the largest contributor to error in terms of magnitude was nonresponse of immunization providers. Total error was most sensitive to declines in coverage due to cell-phone only households.
Conclusions:
Results, which rely heavily on the quality of the estimated components of error, indicated that while response rates and coverage may be declining estimated total survey error was quite small. Since response rates have historically been used to proxy for total survey error, the finding that these rates may not accurately reflect bias is important for the evaluation of survey data.