Gender Turnover and Roll Call Voting in the US Senate

Gender Turnover and Roll Call Voting in the US Senate
of 19
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
  This article was downloaded by: [Bridgewater State College], [Brian Frederick]On: 14 July 2011, At: 15:12Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK Journal of Women, Politics & Policy Publication details, including instructions for authors andsubscription information: Gender Turnover and Roll Call Voting inthe US Senate Brian Frederick aa Bridgewater State University, Bridgewater, Massachusetts, USAAvailable online: 14 Jul 2011 To cite this article: Brian Frederick (2011): Gender Turnover and Roll Call Voting in the US Senate,Journal of Women, Politics & Policy, 32:3, 193-210 To link to this article: 10.1080/1554477X.2011.589281 PLEASE SCROLL DOWN FOR ARTICLEFull terms and conditionsof use: This article may be used for research, teaching and private study purposes. Anysubstantial or systematic reproduction, re-distribution, re-selling, loan, sub-licensing,systematic supply or distribution in any form to anyone is expressly forbidden.The publisher does not give any warranty express or implied or make any representationthat the contents will be complete or accurate or up to date. The accuracy of anyinstructions, formulae and drug doses should be independently verified with primarysources. The publisher shall not be liable for any loss, actions, claims, proceedings,demand or costs or damages whatsoever or howsoever caused arising directly or indirectlyin connection with or arising out of the use of this material.   Journal of Women, Politics & Policy  , 32:193–210, 2011Copyright © Taylor & Francis Group, LLCISSN: 1554-477X print/1554-4788 onlineDOI: 10.1080/1554477X.2011.589281 Gender Turnover and Roll Call Votingin the US Senate BRIAN FREDERICK  Bridgewater State University, Bridgewater, Massachusetts, USA  Most studies looking at the roll call voting behavior of female leg-islators have investigated this phenomenon at the state legislative level and for the US House of Representatives. Very little researchhas looked at the impact of gender on the policy records of US sen-ators. With the number of female senators continuing to increase it is now possible to undertake such an analysis. This study examines the influence of gender in predicting the roll call voting behavior of US senators across several recent congresses. To unearth gender effects, it employs a longitudinal design based on turnover in the Senate, which holds constituency constant while allowing gender and party to vary. The results indicate that male and female sena-tors representing the same state compile very similar voting records on the basic left  / right policy dimension. However, when votes onissues of concern to women are examined, female senators tend to be more supportive than the male senators they replaced, and male senators tend to be less supportive than the female senators they replaced. KEYWORDS women in Congress, Senate roll call voting, gender and representation  As the number of women elected to office in the United States has con-tinued to grow over the past few decades, a range of scholars has tackledthe question of to what extent, if any, does the descriptive representation women of matter? Do female officeholders exhibit distinctive patterns of  Support for this research was provided by the Center for Legislative Studies atBridgewater State University. The author wishes to thank Tracy Osborn and the anonymousreviewers for their helpful comments on earlier drafts of this article. Address correspondence to Brian Frederick, Bridgewater State University, Departmentof Political Science, Summer Street House Office 105, 180 Summer Street, Bridgewater, MA02325. E-mail: brian.frederick@bridgew.edu193    D  o  w  n   l  o  a   d  e   d   b  y   [   B  r   i   d  g  e  w  a   t  e  r   S   t  a   t  e   C  o   l   l  e  g  e   ] ,   [   B  r   i  a  n   F  r  e   d  e  r   i  c   k   ]  a   t   1   5  :   1   2   1   4   J  u   l  y   2   0   1   1  194 B. Frederick  representation on issues that are of substantive concern to women? Muchof the focus of this research has been dedicated to examining whether suchdifferences exist in the roll call voting records amassed by male and femalelegislators. A substantial proportion of this research has concluded that suchdifferences do exist (Reingold 2008). Whether investigating policy represen-tation on a variety of indices of liberalism or conservatism or on specificissues substantively related to the concerns of women, an extensive body of scholarship has documented that women tend be more liberal than theirmale colleagues in each policy domain (Burrell 1994; Clark 1998; Dodson2006; Dolan 1997; Francovic 1977; Rocca, Sanchez, and Uscinski 2008; Swers1998; 2002; Tatalovich and Schier 1993; Thomas 1989; Welch 1985). Whilethis evidence tends to support the proposition that male and female legis-lators have meaningful differences on policy that can be discerned in their voting records, several recent studies have found that women are not signifi-cantly more liberal than their male colleagues (Schwindt-Bayer and Corbetta2004; Hogan 2008). So clearly more work needs to be done to fully flesh outthe contours of this relationship.Even as each of these research studies has made a worthwhile con-tribution to understanding how gender may have an impact on policy representation, there are both methodological and institutional limitationsto what can be gleaned from this collection of scholarship. First, many of these studies may not have properly controlled for constituency effects intheir models. Since women tend to represent more liberal constituencies, any observed differences may be linked to this factor (McDonald and O’Brien2010). Second, most of this research has been conducted in the US House of Representatives, where the institutional design of that body may not be con-ducive to detecting gendered patterns of roll call voting behavior (Frederick2010). The intensely partisan atmosphere in the House has only heightenedover the past few decades, which might explain why many recent studies of that institution have failed to uncover significant gender effects, even when votes of interest to women are examined (Frederick 2009). 1 In contrast, one institution that has not received much attention in theeffort to explore the policy differences of male and female legislators isthe US Senate (Swers 2007; 2008). This absence of systematic inquiry islargely a function of how few women have served in this chamber in itshistory. However, we are now at a point in history where it is time for suchinvestigations to commence. Indeed, at the beginning of the 110th Congressthe number of women reached an all-time high, with 17 women serving inthe US Senate. With its less structured rules and individualistic culture, theSenate is a legislative body well suited for explorations of variation in the voting records of male and female lawmakers at the national level (Frederick2010).This study addresses both the methodological and institutional concernsthat have been raised concerning the findings of past studies. It builds on    D  o  w  n   l  o  a   d  e   d   b  y   [   B  r   i   d  g  e  w  a   t  e  r   S   t  a   t  e   C  o   l   l  e  g  e   ] ,   [   B  r   i  a  n   F  r  e   d  e  r   i  c   k   ]  a   t   1   5  :   1   2   1   4   J  u   l  y   2   0   1   1  Gender Turnover and Voting in the US Senate  195 the previous research by examining the influence of gender in predictingroll call voting behavior in the US Senate over several recent congresses. Todetect gender effects, it employs a longitudinal design based on turnover inthe Senate, which holds constituency constant while allowing for gender andparty to vary. The results indicate that male and female senators representingthe same state compile nearly identical voting records on the basic left / rightpolicy dimension. However, when votes on issues of concern to womenare examined, female senators tend to be more supportive than the malesenators they replaced, while male senators tend to be less supportive thanthe female senators they replaced.  WHY THE TURNOVER MODEL AND WHY THE SENATE? The primary methodological strategies employed in most recent studies test-ing for gender effects in roll call voting have involved estimating a regressionmodel including various measures of congressional voting behavior as thedependent variable with gender as one of the predictor variables, in addi-tion to controls for constituency characteristics including district ideology and other demographic variables (Burrell 1994; Evans 2005; Frederick 2009:2010; McCarty, Poole, and Rosenthal 1997; 2006; Oldmixon 2002; Swers1998; 2002). While some of these studies conduct their analyses for a sin-gle congressional session and others pool data over several congressionalsessions, they are all predicated on the assumption that constituency factors which might be associated with a more liberal voting record are indepen-dent of gender. However, this assumption is somewhat problematic because women tend to be elected in more liberal, more urban, more educated,and more ethnically diverse districts (Palmer and Simon 2008). 2 This real-ity raises the question of whether differences in the legislative behaviorobserved in prior studies is attributable to gender or to the fact that womentend be elected from fundamentally different types of constituencies than domen. Even with an extensive battery of multivariate controls, the regressionmodels cannot account for the fact that the primary independent variableof interest—gender—is correlated with the error term in these regressions(McDonald and O’Brien 2010).One methodological innovation for addressing this concern is theturnover model formulated by Schwindt-Bayer and Corbetta (2004). In theirstudy of the relationship between gender and roll call voting in the USHouse, they control for constituency effects by examining cases of memberturnover in the same congressional district. This approach holds constituency constant without having to introduce an array of district-level explanatory  variables into the regression model. 3 Identifying 167 cases where one Housemember replaced another in the 104th and 105th Congresses, they comparedthe roll call voting record of the member who replaced the previous serving    D  o  w  n   l  o  a   d  e   d   b  y   [   B  r   i   d  g  e  w  a   t  e  r   S   t  a   t  e   C  o   l   l  e  g  e   ] ,   [   B  r   i  a  n   F  r  e   d  e  r   i  c   k   ]  a   t   1   5  :   1   2   1   4   J  u   l  y   2   0   1   1  196 B. Frederick  member and generated an estimate of how much more liberal or conserva-tive his or her voting record was. In their model, they incorporated control variables indicating when a change in party had occurred. Schwindt-Bayerand Corbetta (2004) discovered that party changes produced huge shifts inroll call voting behavior, with turnover from Republican to Democrat produc-ing swings to the left and changes from Democrat to Republican producingmovement toward the right. Yet when they analyzed districts where therehad been gender turnover from a male to a female representative and dis-tricts where there had been a change from a female to a male representative,they uncovered no significant relationship between gender turnover and thechange in the representative’s roll call voting record. These results suggestthat when constituency effects are properly controlled for, it may very wellbe that any differences in the roll call voting records of male and femalemembers of Congress are simply a function of the tendency of women toget elected in more liberal districts. Although the work of Schwindt-Bayer and Corbetta (2004) representsa major contribution to our understanding of gender and roll call votingin Congress, it is still limited in what it tells us about the relationshipbetween these two variables. The first limitation is the dependent variableof interest in their study. Schwindt-Bayer and Corbetta (2004) calculate thedifference in the value of Poole and Rosenthal’s (2007) DW-NOMINATEscore between the new and old member representing the same district astheir outcome variable. While this measure is of great utility in estimatingthe single left / right ideological dimension in Congress, it may not be thebest indicator of whether women and men are voting distinctively on issuesthat are of specific interest to women (Norton 1999). It is possible that menand women in Congress do not vote all that differently when it comes tothe basic liberal-versus-conservative dimension, but they do exhibit distinct voting patterns when it comes to a smaller set of issues that are of impor-tance to women (Reingold 2008). Indeed, past research has shown thatthe relationship between gender and liberalism in legislators’ policy pref-erences tends to be much more robust when women’s issues are examined(Burrell 1994; Dodson 2006; Frederick 2009; 2010; Leader 1977; Oldmixon2002; Poggione 2004; Swers 1998; 2002; Tatalovich and Schier 1993; Thomas1989). Moreover, other studies looking at patterns of bill sponsorship on women’s issues in the House using slightly different versions of the turnovermodel have found that gender is a meaningful predictor of the likelihood of sponsoring such legislation (Gerrity, Osborn, and Mendez 2007; McDonaldand O’Brien 2010). Therefore, before concluding, as Schwindt-Bayer andCorbetta (2004) do, that gender has little explanatory power in understand-ing variation in how legislators cast their votes, it warrants replicating theirmodel using some metric of support for women’s issues as the dependent variable. Another limitation with drawing overly broad inferences based on theresults from Schwindt-Bayer and Corbetta’s (2004) study pertains to the    D  o  w  n   l  o  a   d  e   d   b  y   [   B  r   i   d  g  e  w  a   t  e  r   S   t  a   t  e   C  o   l   l  e  g  e   ] ,   [   B  r   i  a  n   F  r  e   d  e  r   i  c   k   ]  a   t   1   5  :   1   2   1   4   J  u   l  y   2   0   1   1
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