The methodology for this survey was consistent with the process used in compiling prior Agents of Discovery reports. A total of 2,300 surveys were mailed to print and broadcast outlets in every state and the District of Columbia. Each daily newspaper listed in the 2001 edition of the Editor and Publisher Yearbook was sent a survey, regardless of circulation or geographic location. Each licensed television news outlet affiliated with a broadcast network was sent a survey.

The surveys were first mailed to news outlets in January 2002. The survey form, like that of 1999, contained 20 questions and included space for respondents' comments. See a reproduction of the survey form, Appendix A. A cover letter explained the purpose of the study and a pre-addressed, stamped envelope for return of the questionnaire was enclosed. A reminder postcard was sent in April, and in June another letter was sent with a copy of the survey to those who had not yet responded. The Reporters Committee would like to thank Vanessa Leggett, who signed the cover letter of the second mailing urging publishers and broadcasters to answer the survey so that the full extent of the subpoena problem could be documented.

Respondents were asked to provide the number, types and disposition of subpoenas received. They were also asked whether police or other law enforcement officers searched their newsrooms, or if the news organization's telephone records or those of any member of its staff had been subpoenaed during the year. An optional question asked news organizations whether the issuance of subpoenas had affected newsroom policies regarding the retention of unpublished materials or the use of confidential sources.

By the cut-off date, 319 news outlets had responded to the subpoena survey — a 14 percent response rate. See list of participating news organizations, Appendix B. In 1999 the response rate was 19 percent; in 1997, it had been 29 percent.

Of the returned 2001 surveys, 237 (74 percent) came from newspapers and 82 (26 percent) came from broadcasters. The response rate for newspapers was 16 percent, a drop from 22 percent in 1999 and 28 percent in 1997. The response rate for broadcasters was 10 percent, a drop from 13 percent in 1999 and 31 percent in 1997.

The survey also gave each respondent the option of asking that its identity be kept confidential. Of the 319 news outlets that returned survey responses, 117 — or 37 percent — requested anonymity. In this report, organizations that requested anonymity are identified only by state and medium.

The figures and percentages contained in this report have not been statistically analyzed, and no statistical generalizations have been made outside the group of respondents. This report includes only the data supplied by the responding news organizations. Calculations were rounded to the nearest tenth of a percentage point. The sum of the percentages in some questions may not equal 100 percent, both because of the rounding and because in some instances, respondents could select more than one description or answer for one subpoena.