A random sample of 2,400 households was selected using the address-based sampling frame, the U.S. Postal Service Delivery Sequence File. Four contacts were used across all treatment groups, with each contact addressed to the “Resident” of the city or town associated with each postal address in the sample. In order to randomize who responded within households, we asked that the adult (age 19 or over) with the most recent birthday fill out the questionnaire.

This project also included an embedded survey experiment not discussed in this report, which examined issues related to survey sponsorship and mode of request (web or mail).3 Web and mail questionnaire responses are combined for purposes of this brief report. We obtained a final response rate of approximately 49% (n=1104), calculated utilizing American Association for Public Opinion Research standards.