A comparison of ABS mail and RDD surveys for measuring consumer attitudes

Mahmoud Elkasabi

ICF International

Z. Tuba Suzer-Gurtekin and James M. Lepkowski

University of Michigan

Uiyoung Kim

Statistical Research Institute – Statistics Korea

Richard Curtin

University of Michigan

Rebecca McBee

University of Michigan


In consumer research, one of the widely accepted methods to generate a representative sample is random digit dialling (RDD) landline telephone surveys, in which the telephone numbers are sampled from 100-series banks in list assisted sample selection (Casady & Lepkowski 1993). Leading to the nowadays common RDD design, RDD design has passed through many refinements and variants (Waksberg 1978; Tucker et al. 2002; Tucker & Lepkowski 2008). However, the representativeness of the RDD landline telephone surveys is threatened by well-documented non-coverage due to the increasing number of cell-phone-only households in the United States (Blumberg & Luke 2011).