-
Views
-
Cite
Cite
Kyley McGeeney, Courtney Kennedy, A Next Step in the Evolution of Landline Sampling: Evaluating the Assignment-Based Frame, Journal of Survey Statistics and Methodology, Volume 5, Issue 1, March 2017, Pages 37–47, https://doi.org/10.1093/jssam/smw022
- Share Icon Share
While landlines are being eclipsed by cellphones as the dominant sample source in many telephone surveys, they are still commonly used in the United States, as well as internationally. Approaches for drawing nationally representative samples of landline numbers have evolved over time, most notably from the Mitofsky-Waksberg two-stage random digit dialing (RDD) design to list-assisted sampling. This study evaluates a next-generation landline sampling frame designed to address a coverage problem plaguing the list-assisted frame. Using a large, national RDD survey, we estimate the difference in coverage between the list-assisted frame and the new assignment-based frame. We also examine the consequences for bias associated with selecting one landline frame versus the other. The findings indicate that there are systematic differences between landline sample respondents covered by the new frame but not by the list-assisted frame. The overall difference in the coverage rates provided by these two landline frames, however, is so small that those compositional differences have virtually no impact on survey estimates.