Address-Based Area Sampling: An Efficient Hybrid Between Address-Based Sampling and Area Sampling

• Activity Number: 359
• Type: Contributed
• Date/Time: Tuesday, August 3, 2010 : 10:30 AM to 12:20 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Avinash C. Singh and Peter K. Kwok*+
• Companies: NORC and NORC
• Keywords: Address-based Area Sampling ; Design Comparison ; Field Prompter ; Substitution by Rejection Sampling ; Two-phase Questionnaire

Abstract

Use of ABS frame for listing addresses for area sampling (AS) without making field visits is much cheaper, but it cannot be directly applied because some addresses, such as P.O. boxes and rural routes, cannot be linked to housing units (HU). In contrast, cost savings in ABS with mail or telephone interviews (MI/TI) instead of field interviews (FI) in AS are offset by the problems of linking telephone numbers instead of addresses to HU. We propose a new address-based area sampling (ABAS) design whereby a field prompter (FP) is dispatched to verify and update a selected list of ABS addresses, and then drop off short questionnaires to set up a longer follow-up by TI. Ineligible addresses are substituted by rejection sampling techniques developed by Singh and Wolter (2010). We show that ABAS with FP/TI may serve as a good compromise between ABS with MI/TI and AS with FI.