The First Amendment Center, in cooperation with American Journalism Review, commissioned the Center for Survey Research & Analysis at the University of Connecticut to conduct a general public survey of attitudes about the First Amendment. The questionnaire was a national survey developed jointly by the First Amendment Center and the University of Connecticut, in consultation with editors at American Journalism Review. At the University of Connecticut, Chris Barnes, Helene Marcy, April Brackett, Chase Harrison, Katie Stargardter, Professor David Yalof and Professor Kenneth Dautrich directed the project. First Amendment Center Acting Director Gene Policinski and then-executive director Ken Paulson provided overall direction for the project and developed questions with assistance from the FAC's Paul K. McMasters, Charles Haynes, Ron Collins, Brian J. Buchanan and Tiffany Villager. The survey was conducted by telephone between May 6 and June 6, 2004.
Interviews were conducted under the supervision of the Center for Survey Research & Analysis in Storrs, Conn., using a Computer Assisted Telephone Interviewing (CATI) system. All CSRA surveys are conducted by professional survey interviewers who are trained in standard protocols for administering survey instruments. Interviewers assigned to this survey participated in special training conducted by senior project staff. The draft survey questionnaire and field protocols received thorough testing prior to the start of the formal interviewing period. Interviews were extensively monitored to ensure CSRA standards for quality were continually met.
The national sample used for this research project included residential telephone numbers in the 48 contiguous states. The sample was stratified to ensure that broad geographic regions were represented in proportion to their share of the total adult population in the United States. Within each of these regions, telephone numbers were generated through a random-digit-dial telephone methodology to ensure that each possible residential telephone number had an equal probability of selection. Telephone banks that contain no known residential telephone numbers were removed from the sample selection process. Once selected, each telephone number was contacted a minimum of four times to attempt to reach an eligible respondent. Households where a viable contact was made were called additional times. Within each household one adult was randomly selected to complete the interview.
The sampling error for 1,002 national interviews is + 3.1% at the 95% level of confidence. This means that there is less than one chance in 20 that the results of a survey of these respective sizes would differ by more than 3.1% in either direction from the results, which would be obtained if all adults in the appropriate area had been selected. The sample error is larger for sub-groups. CSRA also attempted to minimize other possible sources of error in this survey.