Sample Allocation Using Vendor-Provided Demographic Data

• Activity Number: 540
• Type: Contributed
• Date/Time: Wednesday, August 7, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Survey Research Methods Section
• Author(s): Mike Kwanisai*+ and Kelly Dixon and Dan Estersohn and Alan Tupek and Vrinda Nair
• Companies: and Arbitron and Arbitron and Arbitron and Arbitron
• Keywords: accuracy ; misclassification error ; stratification ; proportional sample ; linear programming

Abstract

In survey sampling initial sample size allocations are generally computed using information such as strata size, response rates and desired target sample size. In practice within strata variability is often unknown so proportional sample allocation is generally preferred. Proportional allocation allows every unit across the different strata to have the same probability of selection and the selected sample units often belongs to the target population. In studies whose target population are Hispanic households, for instance, proportional allocation may yield a sample of households that are not all Hispanic households. The probability of selecting a Hispanic household would depend on the actual number of Hispanic households in the strata and differs by strata. We discuss an allocation approach that uses vendor-provided data about the likelihood of a household unit belonging to the target population. Using linear programming techniques we illustrate how to obtain optimal sample allocation for different strata response and accuracy rates. We also present a sensitivity analysis of accuracy rates obtained from the vendor-provided data.