Address-Based Sampling (ABS) is the sampling of addresses from a database with near universal coverage of residential homes as distinct from the traditional block-listing approach. The use of ABS sample designs gained prominence in the U.S.A. due to concerns with the degradation of the landline sampling frame and the non-coverage of landline numbers in zero blocks, mobile-only households and households without a phone.

The Geocoded National Address File (referred to as G-NAF) produced by the Public Sector Mapping Agency is Australia’s authoritative, geocoded address file that contains more than 13 million Australian physical address records. There is little evidence of this file being used as a sampling frame in Australian social research despite degradation of phone sampling frames similar to what has been experienced in the U.S.A.

The Social Research Centre, has trialled the use of G-NAF as the sampling frame to conduct nation-wide survey as part of Online Panels Benchmarking Study. The sample was drawn on a probability proportional to size basis in accordance with the distribution of the Australian residential population aged 18 years and over across 15 geographic strata (i.e. capital city / rest of State for each State / Territory except the ACT which is treated as one stratum). The trial was conducted simultaneously with two other probability samples (dual-frame CATI and dual frame CATI recruit to online) using the same survey instrument.

The purpose of the presentation is to provide an overview of G-NAF as a sampling frame and to discuss its strengths and limitations based on the findings of the Online Panels Benchmarking study.