Are Female House Members Still More Liberal In a Polarized Era? The Conditional Nature of the Relationship Between Descriptive and Substantive Representation

Are Female House Members Still More Liberal In a Polarized Era? The Conditional Nature of the Relationship Between Descriptive and Substantive Representation
of 23
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Congress & the Presidency, 36:181–202, 2009Copyright  C  American University, Center for Congressional and Presidential Studies ISSN: 0734-3469 print / 1944-1053 online DOI: 10.1080/07343460902948097  A RE F EMALE H OUSE M EMBERS S TILL M ORE L IBERAL IN A P OLARIZED E RA ? T HE C ONDITIONAL N ATURE OF THE R ELATIONSHIP B ETWEEN D ESCRIPTIVE AND S UBSTANTIVE R EPRESENTATION B RIAN F REDERICK  Bridgewater State College Political Science Department, Summer Street House, Bridgewater, Massachusetts Some past studies looking at the voting behavior of women in Congress have shown that they tend to be more liberal than their male colleagues and are more likely to support issues of importance to women. Yet many of these analyses were conducted prior tothe entrance of a number of conservative women into the U.S. House over the past  few election cycles. Focusing on roll-call voting data over 13 Congresses, this studydemonstrates that women in the House are more divided along partisan and ideologicallines than at any point over the past two decades, even more ideologically distant thantheir male colleagues. It presents evidence that over the entirety of this period after controlling for other relevant factors, the effect of gender on roll-call ideology wasstronger for Republican women than for Democratic women. However, in the 108th and 109th Congresses they were virtually ideologically indistinguishable from their male Republican colleagues. A similar pattern has materialized when the analysis is strictlylimited to votes on women’s issues. Thoughgrowingatamodestrate,thenumberofwomeninCongressisatitshighestlevel ever. For instance, at the beginning of the 110th Congress women constituted10.4%oftheHouseGOPConferenceand21.5%oftheHouseDemocraticCaucus. 1 Women occupy key leadership positions in both parties highlighted by the assent of  IwishtothankBarbaraBurrell,CaseyLaFrance,SusanCarroll,SallyFriedman,MicheleSwersandtheanonymousreviewers for their helpful comments on earlier drafts of this manuscript.Address correspondence to Brian Frederick, Bridgewater State College Political Science Department, SummerStreet House, 180 Summer Street, Bridgewater, MA 02325. E-mail: 1 These numbers were taken from the Center for American Women and Politics:  ∼ cawp/Facts/Officeholders/cong-current.html (Last accessed January 26, 2007). 181  182 B. F REDERICK Nancy Pelosi (D-CA) as Speaker of the House. Still, this newfound power is occur-ringatatimewhenthedynamicsofnationalpoliticsarepushingtheDemocratsandRepublicans in Congress farther apart, making bipartisan cooperation an infrequentoccurrence. In the past three decades the U.S. Congress has become an intenselypartisan institution. A plethora of studies have thoroughly documented the rise inpartisanandideologicalpolarizationinboththeHouseandSenatethathasoccurredover this period (Fleischer and Bond 2004; Groseclose, Levitt, and Snyder 1999;Jacobson 2000, 2003 and 2004a; McCarty, Poole and Rosenthal 1997 and 2006;Poole and Rosenthal 1997 and 2007; Rhode 1991; Schickler and Pearson 2005;Sinclair 2006). Congressional members of each party are now more ideologicallyhomogenousandthedistancebetweentheparties’medianpositionsontheleft-rightideological continuum has expanded markedly. Yet only modest attention has beenpaid to the impact of this phenomenon for women serving in the U.S. Congress.For scholars of gender politics the ramifications of congressional polarizationare in the initial phases of investigation. Most, but not all, of the research in thisfield has shown that women legislators tend to display greater liberalism in terms of policy representation (Boles and Scheurer 2007; Burrell 1994; Epstein, Niemi, andPowell 2005; Evans 2005; Swers 1998 and 2002). However, recent changes in thebroader political system at the national level raise significant questions concerningwhether greater numbers of women in office makes a substantive difference in thelegislativeprocessinthefashionthesepreviousstudieshavedocumented.Followingthe 1994 elections, a new collection of conservative Republican women from theSouth and West entered the House taking many positions anathema to traditionalfeminist causes (Dodson 2005, 2006; Evans 2005; Gertzog 2004; Swers 2002;Swers and Larson 2005) 2 Though many women in the House saw these newcomersas “ideological outliers,” their ranks have only continued to expand (Carroll 2002,64). Their beliefs on issues such as taxes, national security, abortion and others aremore in line with the GOP mainstream than Republican women in previous eras.They represent the traditions of economic and social conservatism associated withwomen of the new right (Klatch 1987). In interviews, female members of Congressadmit that intense partisanship and ideological disagreements create an immensestrain on legislative collaboration (Carroll 2002; Dodson 2006; Gertzog 2004).Basedontheemergingevidenceitappearsthatincreasingpartisanpolarizationhas altered the linkage between descriptive and substantive representation. Manyscholars have demonstrated that the political context both endogenous and exoge-nous to institutions in which they serve can deeply influence the ability of womento impact the policy-making process (Dodson 2006; Swers 2002). Therefore, theconnection between the descriptive and substantive representation of women maybe conditioned on the extent to which forces such as the changing electoral basesof the parties exert influence on the context of legislative behavior. Under such a 2 Carroll (2003) finds evidence that a similar trend is occurring at the state legislative level as well.  F EMALE H OUSE M EMBERS IN A P OLARIZED E RA 183 conditional framework, as the constituencies of the congressional parties becomemore polarized the impact of gender in explaining roll-call ideology diminishes.Over the past few decades female Republicans have increasingly represented moreconservative districts and female Democrats have been more consistently electedin very liberal districts. This development has led to the election of Republicanwomen to the House who embrace a more conservative policy agenda than in yearspast and the election of Democratic women whose ideological liberalism divergessharply from their GOP colleagues. Nevertheless, there has not been an extendedlongitudinalanalysisoftheroll-callvotingbehaviorofwomenintheU.S.Congressduring this era of heightened polarization that stretches over a sufficient time frameto empirically test this claim.A glaring need exists to update the literature on whether women in Congressare on average more liberal than their male colleagues, especially in the pastfew Congresses when partisan polarization has reached new heights. In light of this empirical void this study seeks to answer several major research questionsin relation to changing nature of the legislative behavior exhibited by women inCongress in the highly polarized era that has evolved over the past few decades. Towhat extent have women become more polarized over the past two decades? Whatare the electoral bases of this polarization? Are women in both parties still moreliberal than their male colleagues?To answer these questions this study explores the roll-call voting behaviorof women in the U.S. House of Representatives from the 97th through 109thCongresses (1981–2006). It covers the first two years of the Reagan Administrationthrough the first two years of George W. Bush’s second term. 3 Beckwith (2005)argues that there is a dearth of studies tracking the behavior of women in publicoffice across time. The number of congressional sessions covered in this studyallows for a systematic longitudinal analysis to fully explore an important researchquestion in need of additional empirical clarity.Focusing on roll-call voting data, this study demonstrates that women in theHouse are more divided along partisan and ideological lines than at any point overthis period, even more ideologically distant than their male colleagues. Moreover,these rising divisions are linked to the changing nature of the partisan and ideo-logical makeup of the districts female Republican and female Democrats in theHouse represent. The available evidence indicates that over the entirety of thisperiod after controlling for other relevant factors, the effect of gender on roll-callideology and support for women’s issues was stronger for Republican womenthan for Democratic women. However, in the two most recent Congresses, femaleHouse Republicans were no longer ideologically distinguishable from their male 3 The U.S. House is the exclusive focus of concentration since the number of women in the Senate has been toofew to engage in a meaningful comparison of polarization over any extended period. The 97th Congress wasselected as the starting point for the study because it is the first year in Congress that there were at least 10women of both parties in the House.  184 B. F REDERICK Republican colleagues. Ultimately, the findings support the proposition that as po-larization in the electoral bases of the parties expands the probability that greaterdescriptive representation of women will translate into differences in substantiverepresentation declines. This study lends credence to the idea that the link betweendescriptive and substantive representation of women is dependent in large part onthe context of external political forces that shape the institutions where femaleofficeholders serve. PATTERNSOFROLL-CALLVOTINGBYGENDERINTHEHOUSE When examining policy representation in the U.S. Congress, scholars have em-ployed a variety of different measures to ascertain where members stand on theideological spectrum. The most common measure utilized over the past decade hasbeen Poole and Rosenthal’s (1997) DW-NOMINATE scores. 4 Argument supportsthat patterns of voting on women’s issues can best be explained by multiple di-mensions rather than just one (Evans 2005; Norton 1999). After all, NOMINATEscores include procedural votes not explicitly related to the substantive represen-tation women. For most of the analyses presented in this study such an approachis not adopted because as Poole and Rosenthal (1997, 2007) convincingly show, asingleleft-rightdimensionbestcapturesthestructureofroll-callvotinginCongressduring the period covered by this study. 5 However, to rule out the possibility thatgender differences in roll-call ideology may only reveal themselves in the contextof women’s issues, the multivariate models presented later in this study include ananalysis of roll-call behavior on a set of votes exclusively in this domain.Figure 1 plots the mean value of Poole and Rosenthal’s first dimensionDW-NOMINATE scores for female and male Republicans and female and maleDemocrats from the 97th through 109th Congresses. It shows that the trend towardgreater ideological division in the House has not bypassed women. Steadily the av-erage ideological position of Democratic women has grown more liberal whereasthe average ideological location of Republican female House members has grownmore conservative. Two noticeable spikes toward the right for Republican womenoccur during the 104th and 108th Congresses when ideological disputes between 4 These scores are calculated from all non-unanimous roll-call votes in the House and provide an estimate of eachHouse member’s ideology along a unidimensional left-right continuum ranging from − 1.0 to 1.0, with higherscoressignifyingamoreconservativevotingrecord.DW-NOMINATEscoreshaveanadvantageovertraditionalinterest group ratings because they are comparable across time and are complied from a larger sample of votes.However, there are limits to the comparability of NOMINATE scores across time because of the different mixof issues in each Congress (Bailey 2007). 5 Although multiple dimensions have been useful in explaining voting patterns in the House at various points inthe nation’s history, this has not been the case over the past few decades. Members who are more likely to beliberal on the basic left-right continuum are also more likely to be liberal on issues of importance to women.Thus, this measure provides the most effective means to evaluate the extent of ideological and partisan divisionsamong women in the U.S. House.  F EMALE H OUSE M EMBERS IN A P OLARIZED E RA 185 F IGURE 1. Mean DW-NOMINATE Scores by Gender and Party, 97th-109th Congresses. the parties were magnified. In the 97th Congress the mean score for Democraticwomen was about − .34 and by the 109th score this number was about − .46. Thecorresponding figures for GOP women in the House were .22 in the 97th Congress,soaring to .47 by the 109th Congress.The difference in the average position among female Republicans andDemocrats has been on an upward trajectory just as it has in the House as a whole.In the 97th Congress the difference was about .55 and by the 109th Congress thisnumber had reached .94. For most years the difference between female memberswas smaller than it was for males in the House. In earlier Congresses the ideo-logical fissures among women were less pronounced than they were for men. Thiswas the case throughout much of the 1980s when there was a greater percentageof moderate women in both parties. However, in the 108th and 109th Congressesthe difference for female House members exceeded the difference for males. Thegrowing liberalism of House Democratic women and the more consistent conser-vatism of GOP women have finally reached a point where women in the House arenow actually more ideologically distant than their male colleagues.Though women became more ideologically distant they were more liberalon average than their male colleagues. Despite the small sample size of Repub-lican women in the House, closer scrutiny of the data reveals that they exhibitedsignificantly greater liberalism in 8 of the 13 Congresses covered in this study(p < .05). This gender difference in ideology has diminished as in the 108thCongress Republican women were not significantly more liberal than their male
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks