In lieu of an abstract, here is a brief excerpt of the content:

193 I conducted most of my personal interviews of Republican and Democrat representatives and staff, as well as one former Republican House member, in summer 2000. I also conducted a handful in spring 2002. I promised anonymity to all subjects, so their names are not listed. I told interviewees that if they were quoted they would be referred to as “a Republican/ Democrat member/ staffer.” I did not select the subjects at random. I sent letters requesting an interview to all Republican and Democratic members of the Judiciary Committee; the Health, Education, and the Workforce Committee; and the Rules and Appropriations Committees . I selected these representatives because gay rights, abortion , and school prayer-related legislation often go through these committees. I made follow-up calls and conducted interviews with whoever responded affirmatively. I conducted thirty-four interviews. Table A.1 provides the partisan and staff/legislator breakdown. The personal interview is one of several survey research techniques. In deciding which technique to use, researchers face several considerations, such as cost, response rate, and potential bias. Working in a congressional office and knowing the volume of mail received by member offices on a daily basis, I thought self-administered questionnaires (SAQs) would yield a lower than usual response rate. More important, the interviewer has no control over who actually fills out the SAQ; in all Appendix A Elite Interview Information 194 Appendix A likelihood, a staff member would have filled it out. Thus, I proceeded with personal interviews. The survey instrument comprises a series of open-ended questions. I asked all respondents the same questions. My approach to the interviews was more collaborative than standardized , however. I designed the instrument to mimic the social interaction of a conversation rather than the contrived interaction of a traditional interview. For example, when a respondent asked me to clarify the meaning of a question, I did. Some scholars would argue that this technique compromises both the validity and the reliability of the instrument, but I suggest that “what we are looking for is not the standardization of the interaction but stability of meaning across situations and respondents” (Suchman and Jordan 1992, 262). In this case, the benefits of a collaborative interview outweigh the costs. This is a tradeoff, and I placed more importance on “stability of meaning.” Table A.1. Elite Interview Breakdown Interviewee Type Republicans Democrats Staff 7 8 Representatives 13 7 Total 20 15 Note: One former representative is included in Republican totals. ...

Share