The Paradox of Gender Equality: How American Women's Groups Gained and Lost Their Public Voice

Appendix A: Congressional Hearings Data and Other Sources

203 The study utilizes four original datasets of congressional hearing testimony : (1) the Women’s Group Appearances dataset; (2) the Landmark dataset; the (3) Witnesses dataset; and (4) the Transcripts dataset. I assembled and cross-checked each of these datasets by hand from book-length indexes of hearings, transcripts located on microfiche, and the online Lexis-Nexis congressional hearings database. The Women’s Group Appearances dataset contains all appearances by women’s organizations before congressional committees and subcommittees from 1833 (when systematic recordkeeping began) through 2000. The appearances were culled by hand from a review of all volumes of the Congressional Information Service’s CIS Index through 2000. This dataset was cross-checked against hearings data recently put online by Lexis-Nexis. The hearings dataset contains more than 10,400 appearances (the first in 1878) by more than 2,100 groups. To code the content of these hearings, I relied on the Policy Agendas Project, a National Science Foundation–financed project spearheaded by Bryan Jones, Frank Baumgartner, and their colleagues. They have assembled all postwar hearings and assigned each to one of 21 policy domains and one of 228 subdomains. For the postwar hearings, I assigned the codes that Policy Agendas had assigned. For earlier hearings, I used their decision rules to assign the codes. I also coded each women’s group appearance according to whether the witness was testifying on women’s rights, status, or well-being. In some cases, the content was obvious (a feminist group APPendix A Congressional Hearings Data and Other Sources 204 Appendixes testifying at a hearing on the Equal Rights Amendment), but in most cases the content had to be visually inspected by reading the hearing transcript. Finally, I coded the testifying groups by various measures of organization type. This dataset provides the opportunity to chart long-term trends in the volume, agendas, and identities of women’s groups doing policy advocacy on Capitol Hill. The Landmark dataset examines women’s groups’ participation in hearings that informed a subset of 709 particularly important laws (Stathis 2003).1 This dataset, which covers the period 1912–2000, is intended to provide a check on the larger findings. If women’s collective voice is waxing and waning on all nationally prominent issues, is it also doing so on those issues that matter most? The Witnesses dataset provides a second validity check on the Appearances dataset. The Witnesses dataset consists of a random sample of hearings (n = 1,680) at 5-Congress (roughly 10-year) intervals from the late 1870s through the late 1990s. Each hearing was coded for the number and type of witnesses appearing. Witnesses fell into four categories: private citizens , government officials or agencies, corporations, and interest groups. These counts allowed me to calculate whether patterns in women’s groups’ appearances were unique to women or simply reflected larger trends in congressional openness to interest groups or external voices in general. Finally, the Transcripts dataset consists of a carefully constructed sample of women’s group testimony over the 20th century in two policy domains: foreign policy and government provision of health care. The dataset includes 368 pieces of witness testimony from 87 hearings or series of hearings. The transcripts allow for a textual analysis of the shifting rhetorical strategies that women’s groups used to establish their political and policy authority. (For more information about how the Transcripts dataset was compiled, see appendix B.) In terms of understanding organizational change, congressional hearing testimony satisfies three key criteria: availability, reliability, and validity . With respect to availability, thanks to the Congressional Information Service and Lexis-Nexis, I have a complete record of who has appeared before Congress. With respect to reliability, testimony has been performed in a similar fashion for many decades—in person, before House and Senate committee members—and recorded by authoritative sources within the Congress. Most important, testimony has external validity, offering a very good if not perfect measure of women’s democratic voice. Because this book relies heavily on congressional hearings, it is useful to describe what is known about this forum—why groups testify, who gets invited, and what testimony really means. The question of why groups Appendixes 205 testify is neatly summarized by Jeffrey Berry (1997, 164): “The most visible part of an interest group’s effort to influence pending legislation takes place at congressional hearings. . . . Interest group leaders like to testify because it bestows status on them and...