This is the second of a two-paper series presenting an overview of the status of computer-assisted telephone interviewing (CATI) based primarily on published articles and unpublished reports describing experiences with and evaluations of CATI in the U.S. and other developed nations. This second paper examines current knowledge about CATI's consequences for data quality, including effects both on nonresponse and measurement error. It also reviews CATI's new contributions to data collection methodology and suggests needed areas of research to assess better CATI's data collection characteristics.