Telephone Sampling

Martin R. Frankel
Professor of Statistics and Computer Information Systems, Baruch College, CUNY

Abstract
The course will present a mixture of topics on telephone surveys; sampling methods, estimation procedures and operations. It is designed to provide a comprehensive introduction to telephone household and person sampling, covering topics such as the extent to which telephone households represent all households and persons in the US, alternative telephone sampling frames (including the cellular telephone frame), methods of telephone sample selection, estimates for selected telephone sampling methods and features in implementation of telephone sampling methods. The course will cover the history of telephone sampling in the US. It will focus on the basic methods of list-assisted random digit dialing (RDD) telephone sampling that are in general use. It will also deal with more advanced topics such as dual frame telephone sample designs, geographic and demographic over-sampling as well as the use of interruption and propensity weighting methods to adjust for potential bias due to non-telephone households. Participants will be given a brief overview of operational issues such as call scheduling and the use of sample replicates and the computation of various response rates.

Instructor
Martin R. Frankel is Professor of Statistics and Computer Information Systems at Zicklin School of Business, Baruch College, CUNY. He works with a number of research organizations including Abt Associates and Mediamark Research. He has worked in the area of telephone surveys for more than 30 years. His research encompasses telephone sampling design, implementation and estimation. He has served as a consultant for public and private organizations including, Rand Corporation, NBC Universal, NYCPD Forensic Laboratory and the Motion Picture Association of America.