Representing women voters

Abstract

While analysis of the relationship between the descriptive and substantive representation of women has mainly focused on women politicians as critical actors in many contexts political parties provide the linkage between voters’ preferences and policy programmes. The manner in which political parties respond to women voters is shaped by both the information they receive about women voters’ preferences, from the news media, pollsters and other sources, and by gendered party type. Analysis of parties’ attempts to target women voters can help us understand whether parties perceive women as typical or average voters. I conduct two case studies on the influence of gendered newsframes on party policy and inter-party competition in the United States and Great Britain to formulate a preliminary analytic framework designed to facilitate research to assess how parties’ responses to portrayals of women voters vary according to institutional and contextual factors across time and space.

Keywords

gender gap newsframes voting behaviour

‘When we’re dealing with a gender-gap issue, there are two ways of looking at it – superficially through a public relations campaign or substantively through legislation’ Congresswoman Schneider (cited in Abzug, 1984: 138)

Introduction

The core intention of this article is to provide a preliminary analytic framework capable of comparative application, to apply to the question, ‘is the electoral linkage between political parties and women voters deep-rooted and central to electoral campaigns, or, on the contrary, superficial, ad hoc and volatile?’ Two illustrative case studies are used to set out an innovative research agenda for extensive, in-depth, collaborative work to compare the response of parties – across time and space – to gendered representations of voters.

The majority of the extensive research undertaken by feminist political science on the link between the descriptive and substantive representation of women has focused on the role of women/feminist politicians as critical actors (Celis and Childs, 2011; Celis et al., 2008; Childs, 2007; Dovi, 2002, 2007; Mansbridge, 2003; Phillips, 1995; Wängnerud, 2000). However, in many contexts, political parties provide the linkage between voters’ preferences on the one hand and policy programmes on the other; critical actors serve as intermediaries who attempt to convince party elites to incorporate women’s interests into their policy programmes. Critical actors make representative claims with regard to women (Celis et al., 2014; Saward, 2010) often by employing polling data to produce an account of women voters’ preferences. The extent to which these interventions are successful may vary depending on the configuration of a number of factors: such as the mobilisation of feminist (usually) women in the legislature and the upper echelons of the party hierarchy; the ideological position of the party and the strength of women’s/feminist organizations in civil society (Kittlison, 2006; Schwindt-Bayer and Mishler, 2005; Young, 2000). Another factor that is likely to have a significant impact on the ability of feminist actors to influence parties is the electoral context.

In an environment where women voters’ behaviour is, or is perceived to be, distinct from men’s parties may be more responsive to demands to represent women’s interests if they anticipate gaining an electoral advantage. For example in the United States and Britain when women’s suffrage was first granted political parties feared that women would vote as a block and they responded by developing policy programmes designed to be attractive to women (Burrell, 1993: 301; Carroll, 2010: 123). The stream of ‘women friendly’ policies however dried up when it became evident¹ that women differed little from men’s (Abzug, 1984: 30, 109; Mueller, 1988: 20; Norris and Lovenduski, 1993).

In the early 1980s, evidence of partisan dealignment (Nie et al., 1979) demonstrated that many of the previously stable political identities (Lipset and Rokkan, 1967), that parties had targeted in order to secure an electoral mandate, had become less politically salient across advanced industrial democracies. The fracturing of these old political identities, and the rise of new politics in the 1960s and the 1970s, provided a window of opportunity for feminists to renew their claims that there was an electoral incentive for parties to take women seriously (Norris, 2001: 9–10); and many members of the women’s movement switched tactics and began to participate in electoral politics. As a result, we see further evidence that male-dominated parties respond to women voters when they fear that they may vote as a block, in their reactions to the emergence of a ‘modern gender gap’ in the United States.

The gender gap is part of an enormous wave of demographic change sweeping the country that threatens to swamp the Republican Party. Edward J. Rollins, President Regan’s chief political advisor (Cited in Abzug, 1984: 3).

The greatest political upheaval of the century [the gender gap] Patrick Cadell polling analyst for the Democrats (Cited in Abzug, 1984: 4).

Thus, where there are gender gaps in political attitudes and behaviour, there may be an opportunity for feminist actors to exploit these gender gaps to secure policy gains. It is well documented that feminist activists have used portrayals of women voters as leverage to pursue feminist goals (Norris, 2001). In the United States, the high-profile modern gender gap, evident since 1980, where more women than men support Democratic candidates, has been utilized by feminists to support their case for pursuing a feminist conception of women’s interests. In the United Kingdom, the Labour Party’s adoption of all women shortlists (1993) in the selection of parliamentary candidates has been, at least partially, accredited to a sustained campaign undertaken by women in the party to draw the leadership’s attention to the historic advantage the Conservative party had among women voters (Eagle and Lovenduski, 1998; Russell, 2005). If the supposition, that high-profile gender gaps in political attitudes and behaviour provide fertile ground for feminist policy initiatives, is correct then where gender gaps become salient features of media coverage of electoral contests we would expect to see the parties competing to capture gendered votes.² The extent to which the parties are themselves feminised is likely to influence whether any reaction to media portrayals of women voters is either fully integrated into a party’s offer or superficially appended to it. Additionally exogenous shocks to the party system may open up windows of opportunity where feminist actors are able to direct the party’s attention to the perceived interests of women voters (Childs, 2015: 9).

The feminisation of political parties can take place on two levels: deep-rooted and superficial. Deep-rooted feminization occurs when women’s organizations, interests and perspectives are embedded within party structures and routinely incorporated into policy platforms and campaigns. Superficial feminization occurs when women’s interests are accommodated in an ad hoc and piecemeal fashion in response to a perceived electoral case on an election by election basis. Figure 1 represents a schematic approach to classifying gendered party types along two dimensions: gender ideology (ideas) and the integration of women within the party (numbers; à la Lovenduski, 2005). A party with a liberal feminist ideology will include the substantive representation of women among its core beliefs and values. As such, a party with a liberal feminist ideology should be highly responsive to feminist portrayals of women voters. It is not necessarily inevitable that a party with a feminist ideology will also include large numbers of feminist (usually) women integrated within the party,³ but the two dimensions are highly correlated because feminist activists working within the party over a long time period are likely to influence its core beliefs and values. Thus, a feminized party is most likely both feminist in its ideology and within its party structures. I hypothesize that feminist claims regarding women voters will provoke a more deep-rooted response from a feminized party than one with a traditional gender ideology and few feminists (usually) women mobilized within the party hierarchy.

Figure 1.

Gendered party types, United States and United Kingdom two largest parties.

Case studies

In order to highlight the need for an analytic framework to assess the relationship between feminist activism, media newsframes and party behaviour I employ two case studies (the United States and Great Britain). The criteria for case selection are that in both countries, the issue of gender became a salient feature of the election campaign coverage (from 1980 in the United States and 1997 in the United Kingdom). Focusing on these high-profile cases maximizes the likelihood that parties will be aware of, and therefore, possibly responsive to, gendered newsframes either at a surface or deep level. Future research should attempt to explain party response by capturing variation in media coverage and feminist activity to establish what factors determine party response across time and space (as set out in the preliminary analytic framework, Table 1).

Table 1.

Proposed analytic framework.

Dependent variable/s

Party liberal feminist campaign scale comprising of analysis of:

Party manifestos

Leaders’ speeches;

Election policy commitments and

Campaign materials.

(Gender gap in party of vote)

Independent variables

Party feminization dimension 1:

Proportion of women representing the

party in the:

Legislature;

Executive and

Mass membership.

Relative power of party women’s

organizations

Party feminization dimension 2:

lite survey rating parties’ position on a liberal feminist scale over the previous legislative period.

Feminist campaign activity outside of the parties

Gendered voter newsframes:

Frequency of articles referring to women voters;

Frequency of use of gendered newsframes;

Feminist/traditional gender ideology driving the newsframe (Y/N) and

Newsframe created by feminist activists (Y/N).

Newsframe created by anti-feminist activists (Y/N).

Newsframe created by party (Y/N).

Additional information:

Newsframe supported by sound empirical evidence (Y/N).

Actual size of gender gap at last election

The United States

The US gender gap literature is extensive, evolving and highly sophisticated. In this case, study I uses the gender gap literature to identify gendered newsframes employed during election campaigns. I trace the circumstances in which a newsframe gains purchase and identify causal factors that should be isolated in a systematic comparative analysis across space and time. The high-profile ‘modern’ gender gap in the United States has generated a massive body of research that has influenced analysis of gender and voting behaviour across the globe. The gender gap is not a chasm (Seltzer et al., 1997), but it has been relatively stable since 1980, remaining between 4 and 11 percentage points (Figure 2).

Figure 2.

US gender gap in presidential elections 1980–2012.

In the United States, little public attention was given to potential differences between men and women voters – after the bout of consternation at the time of women’s suffrage – until the gender gap was revealed in the aftermath of the 1980 presidential election, where 8% fewer women than men voted for Ronald Regan (Frankovic, 1982; Sigel, 1999).⁴ In fact, there was a small gender gap in 1976, with more women supporting Carter, but this fact did not receive media attention and was dismissed by Carter’s own pollster as unimportant (Mueller, 1988: 17). It took feminist intervention for the modern or left-leaning gender gap in the United States to gain public attention. The first use of the term gender gap is attributed to Eleanor Smeal, then president of the National Organization for Women (NOW) in 1981 (Bonk, 1988: 89; Sigel, 1999: 5). Roberta Sigel (1999) argued that by naming the phenomena – gender gap – and providing a coherent narrative about women voters the women’s movement was able to draw politicians’ attention to the issue of the representation of women (Sigel, 1999: 5).

The naming of the gender gap kept the issue of women’s interests and concerns on the political agenda and encouraged the parties to consider women candidates; for example, it influenced the selection of Geraldine Fararro as the Democrat candidate for Vice President in 1984 (Norris and Carroll, 1997: 9). Carol Mueller argues that NOW effectively utilized the gender gap to convince the Democrats that including a women as the vice presidential candidate would be electorally advantageous (Frankovic, 1988; Mueller, 1988). However, both CBS and the New York Times produced polls that showed that including a woman on the ticket would have no effect on the result but neither polls were published, as they did not fit with the prevailing newsframe (Mueller, 1988: 29). NOW’s campaign to maintain a feminist newsframe was successful in the short term because of the resources they deployed and because their frame fulfilled the news media’s demand for exciting news stories, with a coherent narrative, as a result the gender gap frame influenced the Democrat’s decision to choose a woman to run for vice president (VP). Thus, feminist representations of women’s interests can be demonstrated to have had a significant impact on the party’s decision-making (directly improving the descriptive representation of women in the form of a woman vice presidential candidate), but in this case, success was not driven by the use of reliable evidence regarding the demands of women voters. Instead, the narrative that women would be more inclined to vote for a party with a women running as VP was not supported by the opinion poll data, but evidence that ran counter to the newsframe was excluded from media portrayals of women voters. This example illustrates why a comprehensive account of how parties represent the interests of women voters requires the inclusion of an analysis of how information about women voters is conveyed to the parties rather than simply linking survey data to party policy without considering the intervening variables of feminist activism and media content.

The gender gap newsframe, although still in use, was accompanied by the Year of the Woman in 1992, focusing on the record numbers of women running for and elected to Congress (Dolan, 1998). Sue Carroll contends that although women’s organizations, such as NOW, initially benefited from the surge in attention to women voters in the Year of the Woman, over the longer term interest may have waned as a result of media fatigue and interest in women candidates declined in 1994 and 1996 (Carroll, 1999). Attempts to estimate the influence of the women’s movement on the parties’ attitudes toward women voters should include a measure of the fluctuation in campaign intensity (in this case as a result of diminishing resources available to NOW (Bonk, 1988; Mueller, 1988) to assess whether, as feminist activism drops off, other actors’ framing of women’s interests become dominant.

The Year of the Woman was followed by the Year of the Angry White Male in coverage of the 1994 House elections. The shift to a focus on men resulted from an opinion poll showing that men had transferred their support to the Republicans, as a result women were largely absent from media portrayals of voters and candidates during the 1994 election (Carroll, 1999: 8). The shift of emphasis from women to men was criticised on the grounds that it simply reflected the media’s appetite for a new angle rather than a response to credible evidence of a shift in the behaviour of men.

Scholars with access to the National Election Studies have searched hard for evidence about the angry white male in the 1994 elections, but they have come up empty-handed (Norris and Carroll, 1997: 11).

Yet more evidence that media portrayals of voters should be included in models of parties’ attempts to represent women voters.

The 1996 presidential election saw the arrival of the Soccer Mom. The term was cited by E. J. Dionne in The Washington Post on 21 July, 1996 when he quoted a senior media adviser to Bob Dole, Alex Castellanos, who said that Clinton was following the advice of his pollster, Dick Morris, and targeting a voter whom Castellanos called the ‘Soccer Mom’ (Norris and Carroll, 1997); ‘the overburdened, middle income working mother who ferries her kids from soccer practice to scouts to school’ (Dionne Jr, 1996). Clinton responded to the Soccer Mom frame by speaking on issues that his team believed would appeal to this specific group of target voters, such as school uniforms and curfews for teenagers (Carroll, 1999: 10). Dole’s response was more explicit than Clinton’s and he referred to Soccer Moms directly on several occasions and argued that his proposed tax cut would allow more women to stay at home with their children rather than seek paid employment (Carroll, 1999: 10). To some extent, neither candidates’ responses to the Soccer Mom frame were liberally feminist as they focused almost exclusively on children rather than the core demands of the organized women’s movement – such as the liberalization of access to abortion, access to healthcare or childcare or other feminist concerns, such as sexual harassment (ibid). The Soccer Mom frame resulted in an emphasis on policies that impact on children and women were used as a proxy for family concerns. In fact, in her analysis of the same election, Sue Carroll concluded that the widespread use of the gendered Soccer Mom newsframe may have been detrimental to the ability of women’s organizations to translate the media and party interest in women voters into tangible gains for women (Carroll, 1999). This example again shows that models of party behaviour should include competing representative claims made about women voters and should incorporate the dominant narrative evident in any given election cycle.

However, a small number of the Clinton ads did cover feminist concerns (Jamieson et al., 1999) and Clinton referred to women in his speeches and made use of them as examples more often than Dole did. Clinton’s even suggested God might be a women (Jamieson et al., 1999: 14). This analysis demonstrates that despite Carroll’s critique that Clinton’s attempts to target women voters were overly narrow they were considerably more liberally feminist than Dole’s overtures. This is in keeping with a feminist analysis of Democrat and Republican gendered party types, which demonstrates that even in the context of a high-profile gender gap gendered party type plays a crucial role in the extent to which campaign messages and policy platforms are designed to meet the interests of feminist women.

The potential for dissonance between gendered newsframes and the empirical evidence is manifest in the post-election analysis of the 1996 presidential election.

Ultimately, … the soccer mom did not play a particularly important role in Clinton’s victory. Like the “angry white male” before her, on further analysis, systematic evidence establishing the importance of the soccer mom proved elusive. (Kaufmann and Petrocik, 1997 cited in Norris and Carroll, 1997: 12)

It would seem that although the Soccer Mom newsframe did direct some attention to women voters, it shifted the media and parties’ attention from the feminist framing of the gender gap and the need to address a wide range of ‘women’s issues’ to a much narrower conception of women’s interests which elided women with the family. Hence, it is important that any analysis of the influence of gendered newsframes on parties’ attempts to represent feminist women voters includes an assessment of the extent to which the newsframe is liberally feminist.

The Soccer Mom reappeared, although to a lesser extent, in the 2000 presidential election campaign and she was followed by the Security Mom in 2004 (Carroll, 2006, 2008: 77). The Security Mom had much in common with the Soccer Mom: she was perceived to be white, married with young children, but in the wake of the 2001 terrorist attacks, she was extremely anxious to protect her family (Carroll, 2006: 93). And, much like the Soccer Mom, ‘empirical data collected from voters offer little support for the characterization of security moms as portrayed in media accounts’ (Carroll, 2008: 76). Kerry and Bush both addressed the Security Mom by stressing that they would be tough on terrorism, but Kerry also focused on healthcare and the gender pay gap in an attempt to gain support among poorer women who were over-represented in the undecided category. Both candidates made symbolic attempts to appeal to women voters by making public appearances with prominent women and in addition, Kerry made some targeted issue-based appeals. Sue Carroll concludes that the Security Mom frame benefited the Bush campaign because it allowed him to appear to address women’s concerns without making any tangible policy commitments (Carroll, 2008: 87). Again the Security Mom frame deflected attention away from other subgroups of women, from the women’s movement and from a feminist reading of the gender gap (Carroll, 2006). Although the use of the frames meant that women voters received some attention during the campaign, it would seem that neither the Soccer Mom nor the Security Mom were likely to encourage the parties to attempt to represent liberally feminist women.

In the 2008 and 2012, presidential elections women voters received more media attention than ever before (Carroll, 2010, 2014). In 2008, Hillary Clinton’s candidacy in the Democratic primaries and the high levels of support she received among women ensured that women voters were a central part of the media’s coverage of the election (Carroll, 2010). Again the candidates made symbolic appeals to women voters by being seen with high-profile women and the Republicans sought to push back against the Democrats’ advantage among women by selecting a woman, Sarah Palin, to run as VP (Carroll, 2014). The selection of a woman to run as VP was McCain’s most serious attempt to target women voters in 2008. Obama made more issue-based attempts to attract women voters, stressing economic initiatives to help working women (Carroll, 2014: 120). In 2008, there was not a dominant gender newsframe and instead women voters were portrayed in a variety of different ways and they were central to the election coverage. In 2012, the most prominent gender newsframe was generated by the Obama campaign itself; they developed a narrative of the Republican War on Women. The narrative had purchase because of the extreme and shocking statements made by a number of Republicans during the primaries and the campaign: including Republican Senator Rick Santorum saying that contraception is ‘not okay’ and Republican Congressman Todd Akin stating that women’s bodies prevent them from getting pregnant as a result of rape (Carroll, 2014: 121–122). In addition, Obama repeatedly stressed his support for equal pay for men and women, while Romney publicly dodged the question. Romney’s campaign made little reference to women’s issues and in contrast, Obama frequently raised issues of women’s health and pay equality. It is clear that gendered newsframes often originate from within political campaigns/organizations (both feminist and non-feminist) and critically the extent to which they are liberally feminist in content is likely to both reflect and shape the parties’ attempts to represent women voters.

It would seem that 32 years after NOW publicized the gender gap, the story of how women voters are portrayed by the media and represented by the parties in the United States is one of considerable variation and change over time. The extent to which parties respond to women voters in an ad hoc or more systematic and holistic manner does seem to be influenced by the way women voters are portrayed in the campaign and which actors are most successful in shaping that narrative.

Great Britain

Historically women in Britain voted for the Conservative party in greater numbers than men (see figure 3) (Norris and Lovenduski, 1993: 38). In the United Kingdom, feminist activists within the Labour Party in the 1980s and early 1990s were successful in linking this gender gap in voting behaviour to the under-representation of women in Parliament and this narrative played a critical role in the party’s decision to adopt a gender quota, in the form of all women shortlists, in the selection of parliamentary candidates in the run up to the 1997 General Election. The use of women voters to argue for the better representation of women in politics was set out in two pamphlets published by prominent Labour women (Brooks et al., 1990; Eagle and Lovenduski, 1998; Hewitt and Mattinson, 1987). The attempt to secure the better representation of women voters was largely an internal party campaign and not driven by media-generated gendered newsframes. However, the party’s attempts to target women voters did lead the media to identify newsframes to pitch the story.⁵

Figure 3.

The British gender gap 1945–2010 (adapted from Norris, 1999).

Feminist advocacy groups were particularly active in the 1997 election campaign, producing reports and polls and generally attempting to push the issue of women voters onto the media agenda and gaining considerable coverage (Lovenduski, 1997). The press coverage did not include one single-dominant gendered newsframe and both feminist and anti-feminist portrayals of women voters were in evidence (Lovenduski, 1997: 714). However, in 1997, Worcester Woman was identified by Conservative pollsters and adopted by the media as a newsframe to describe working-class women in marginal constituencies with children who might have previously voted Conservative but who had switched to Labour. To some extent, Worcester Woman was borne out of the Labour feminists’ campaign for the party to take women seriously by drawing attention to the traditional gender gap, but it also narrowed the focus to women with children. In addition, women MPs were subject to a gendered newsframe after the election in the form of Blair’s Babes, which was prevalent in coverage of the increase in the number of Labour women MPs from 37 to 101. Again there was not a dominant gendered newsframe during the 2001 election campaign, although Worcester Woman returned, and women journalists took up feminist criticisms of the absence of women politicians from key campaigning roles (Lovenduski, 2001). There was evidence that the parties were seeking to target women’s votes, most prominent was the Labour Party, which produced a manifesto for women, but the Conservatives also produced some policies targeted at liberally feminist women (Lovenduski, 2001: 748).

In the run up to the 2005 General Election, the parties and the media were unusually interested in women voters. The media attempted to define women voters by applying gendered newsframes, Worcester Woman returned again, accompanied by Let-down Woman and Do-it-all woman. Let-down Woman reflected a belief that women were especially unhappy about Britain’s involvement in the Iraq war and that they were disaffected with Tony Blair, supported by a poll commissioned by the liberally feminist advocacy group the Fawcett Society. However, women’s supposed disillusionment with Blair was probably an artefact of poor research methods; reports of polling evidence that women were unhappy with Blair frequently did not offer a comparison with men and implied a gender difference that was not in evidence. The Let-down Woman frame created the impression that women were moving away from Labour when in fact there is no evidence to support this interpretation. The focus on women’s alleged hostility to Blair also suggested that women respond to party leaders in a more personalized and emotional way than men – headlines asked questions, such as whether ‘women had fallen out of love with Blair’ and discussed attempts to ‘woo’ women voters – again a stereotype without empirical support. All three parties used the media to engage in direct appeals to women voters. Labour and the Liberal Democrats produced manifestos for women and Labour women filled a bus with women MPs and visited marginal constituencies to speak to women voters, although the bus received little media attention (Campbell and Lovenduski, 2005: 847; Childs, 2005). The Conservative Party’s third consecutive defeat in 2005 led them to elect a new party leader, David Cameron. Cameron sought to detoxify the party’s brand by portraying the party as being in touch with modern Britain: A key part of the strategy was party feminization (Childs and Webb, 2012). Cameron chose to make his first speech as party leader at the Equal Opportunities Commission and he focused on women’s issues from a liberally feminist conception of women’s interests. As a result, from 2005 onwards, the difference between the main parties on feminist ideology diminished considerably.

In the British news media, 2010 was going to be the ‘Mumsnet Election’. The term was coined by Rachel Sylvester in the Times (17/11/09) and by 5th May, 43 further articles in British national newspapers used the term. Sylvester claimed that the election ‘would be decided at the school gate’ and she depicted the party leaders as lining up to participate in live web-chats with women voters through the Mumsnet site. The focus on mothers was confirmed by Douglas Alexander, the Labour Party’s 2010 election coordinator, when he said that ‘Labour needs to win back middle-income female voters with children in marginal seats’ (Guardian, 19.2.10). The Mumsnet Election was a significant feature of the election newsframe that set up a narrative that women voters, and policies that affect women more than men, would be at the centre of the campaign. The three main parties did set out policies specifically targeted at women and mothers in particular, work/life balance issues were addressed in all three election manifestos with promises to make maternity and paternity leave more flexible. All three parties were also committed, rhetorically at least, to improving the representation of women in Parliament. All three main parties offered a liberally feminist set of policy propositions and although the Labour Party provided the most comprehensive offer for liberally feminist women, the Liberal Democrats and the Conservatives actively competed to secure their votes.

During the 2015 British General Election, women voters became a salient feature of the election coverage when the Labour Party launched its pink battle bus intended to reach out to women voters. The bus generated a great deal of publicity, not because the party was targeting women voters per se but due to its colour; generating many column inches that at least touched on the issue of women voters. The Labour Party were accused of patronizing women and of succumbing to the ‘pinkification’ of womanhood; although – as previously noted – the Labour Party had employed campaign buses (of other hues) seeking out women voters in previous elections which had been largely ignored by the media. Admirable as the attempt to draw attention to women’s issues may have been, the bus’s deployment was largely publicly justified on the basis of dubious statistics. The Labour Party claimed that there were 9.1 million missing women voters in the 2010 British election whereas in fact there was not a statistically significant gender gap (Campbell et al., 2015). The ‘missing’ women voters became a recurrent news story and thus a good deal of the media coverage and public debate about women voters that took place was based on the erroneous belief that women had been less likely to vote in 2010 than men. This is another example of a representative claim made about women voters that became part of the news agenda but was not supported by sound evidence. As a result in order to understand the parties’ attempts to target women voters during the 2015 election, we need to understand how they responded to this portrayal of women as disproportionately non-voters as well as including observational data about ‘real’ gender differences in analytic models. Future research should assess whether the other parties responded to the frame and offered feminist policies in order to attempt to compete for women’s votes as a result.

Although there wasn’t a single coherent gendered newsframe in play during the 2015 election, a Nexis search of UK national newspapers shows that considerable newspaper copy was devoted to the subject. In fact during the final 3 months prior to the election in 2015, there were 142 references to women voters in national newspapers compared with 45 references in the same period in 2010. There was not an easily digestible simple newsframe such as the Mumsnet Election or ‘Worcester Women’ to provide a journalistic hook for articles discussing women voters, but women voters and women’s issues were raised frequently. It is possible that the issue of women voters was more fully integrated into the coverage of the 2015 campaign as a whole. Analysis of the parties’ manifestos shows that all of the parties offered liberally feminist policy positions on issues ranging from childcare to domestic violence and even UKIP, the least feminist party, put forward a set of policies designed to attract liberally feminist women voters (Campbell et al., 2015). The 2015 British General Election was highly gendered in other regards, the prominence of the small parties in the leaders’ debates in the election coverage (three of the seven party leaders were women), meant that women politicians were more visible in the campaign than in previous years. Further research should assess whether an electoral context where women voters and women politicians, framed from a liberally feminist perspective, are highly visible results in inter-party competition for women’s votes.

To sum up, there is no longer a statistically significant aggregate level gender gap in voting behaviour in the United Kingdom. Women voters are a recurrent feature of election coverage and there appears to have been an upsurge in the attention the media pays to them. Crucially all of the main political parties have adopted liberally feminist policy platforms, particularly since the election of David Cameron as Conservative party leader; the main parties are explicitly competing for the votes of liberally feminist women. In this context, it is perhaps not surprising that a modern gender gap (Inglehart and Norris, 2000) has not emerged in Britain. There is less variation in gendered party types in the United Kingdom than in the United States, which may help explain the variation in the gender gap across the two countries. Returning to our central research question, there has been a significant movement in the attempts of the main parties in the United Kingdom to incorporate liberally feminist policies into their election commitments. The application of the analytic framework, set out later, would help us to unpick whether this change has been driven by observable shifts in women’s preferences, feminist activism, gendered newsframes or all three.

The analytic framework

In this section, I use the insights gained from the United States and United Kingdom case studies to elucidate a preliminary analytic framework to guide future comparative research. In order to provide a convincing answer to the research question, ‘is the electoral linkage between political parties and women voters deep-rooted and central to electoral campaigns, or, on the contrary, superficial, ad hoc and volatile?’ researchers must operationalize a series of concepts.⁶ These concepts include gendered party types (measured using the two dimensions of party feminization), feminist and anti-feminist campaigning from inside and outside of the parties and gendered newsframes. The extent to which election campaigns address liberally feminist concerns can be assessed through the use of content and discourse analysis of party campaign materials, manifestos and speeches. In order to generate comparable data, country researchers should develop a common coding frame and create a cumulative party liberal feminist campaign scale generating a score for each major party seeking election. The first dimension of feminism could be operationalized by collating information about women/feminists representation throughout the hierarchy of the political parties. The second dimension of feminism could be operationalized by conducting a survey of gender and politics scholars to place the parties on a liberal feminist scale over the previous legislative period.

The extent to which gendered newsframe/s are evident in election coverage could be captured with a simple frequency score for the number of references to women voters subdivided by the keywords employed in the gendered newsframe/s in coverage of the campaign. For the purposes of this analysis, gendered newsframes refer to media portrayals of women voters that have a single coherent narrative and a frequently employed phraseology, such as gender gap, Soccer Mom or Security Mom (Jamieson and Waldman, 2004). The gendered newsframe may be coded as liberal feminist or anti-liberal feminist and may originate within feminist organizations (inside and/or outside the parties) or from outside of the women’s movement. Undertaking this data collection exercise would require the creation of a network of country researchers and the deployment of substantial research resources, but the potential for significant advances in our understanding of how/when the substantive representation of women occurs and, potentially, the sources of the gender gap itself are sizeable. Collating detailed and systematic comparative data, as set out in Table 1, would enable a comparative analysis of how political parties respond to liberally feminist representations of women voters over time and place that would incorporate variations in election context and political activism.

Conclusion

A historical reading of how parties have been persuaded to address women’s issues reveals that a case that women’s votes depend on it is very commonly made. Critical actors are important but they are most successful when they are able to claim that they can bring women voters with them. The study of gender and voting behaviour and gender and party politics form two distinct strands of gendered political research but they should be more fully integrated. This is a real challenge because it requires extensive collaborative research across time and space, but such an approach would provide more complete accounts of how the representation of women occurs.

It is time to bring politics back into models of the gender gap. The US case study demonstrated that the 1980 US presidential election was a critical election in gendered terms as it was the first where a modern gender gap was apparent. The year 1980 also possibly represents a critical moment in terms of the polarization of the two main parties on gender ideology (Freeman, 1987; Gillion et al., 2014; Rymph, 2005). The successful Republican presidential candidate, Ronald Reagan, was the most traditional in his view of gender roles among his Republican rivals; stating that a wife in the workforce ‘threatens the very structure of family life itself’ (Abzug, 1984: 81–87) and arguably this was the moment where the two parties’ gender ideology clearly diverged or arguable reversed (Freeman, 1987). Early studies of the gender gap incorporated the behaviour of the parties into their analysis of the gender gap; for example, Sue Carroll described Ronald Reagan as a ‘catalyst for the gender gap’ (Carroll, 2010: 125), but most recent studies focus on sociological explanations for differences in men and women’s voting behaviour. Most current analysis of the gender gap in voting behaviour treats the party positions as relatively stable, but if 1980 was a critical moment then there is reason to measure the parties’ gender ideology and mobilization of women within the party hierarchy over time to assess the impact this has on women’s support for the parties themselves. In these circumstances, we would expect the Democrats to be more responsive to feminist portrayals of women’s policy interests and to prove more attractive to liberally feminist women voters than the Republicans, arguably a significant source of the US gender gap. In the United Kingdom, however, the main parties are much closer together in terms of gender ideology, particularly since 2005 (Campbell et al., 2015). In this context, we may not expect to see a gender gap in voting behaviour as the main parties are competing to capture ‘feminist’⁷ women’s votes. Thus, understanding gender and voting behaviour in global perspective requires an account of both the sociological factors shifting gendered attitudes and the reactions of political actors.

Throughout the case studies, it was clear that political parties create and react to gendered newsframes and that the extent to which they respond by incorporating feminist demands depends on gendered party type and the feminist (or not) nature of the newsframe itself. It is important, therefore, that any academic analysis of the relationship between gender gaps and the descriptive and substantive representation of women, as facilitated through political parties, includes measurements of both actual gender differences in political attitudes and behaviour and media portrayals of these differences.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Abzug

(1984) Gender Gap: Bella Abzug’s Guide to Political Power for American Women. Boston: Houghton Mifflin Company.

Bonk

(1988) The selling of the ‘gender gap’: the role of organised feminism. In: Mueller

(ed.) The Politics of the Gender Gap: The Social Construction of Political Influence. Newbury Park: Sage, pp. 82–101.

Brooks

Eagle

Short

(1990) Quotas Now: Women in the Labour Party. London: Fabian Society.

Burrell

(1993) Party decline, party transformation and gender politics: the USA. In: Lovenduski

Norris

(eds) Gender and Party Politics. London: Sage, pp. 291–308.

Campbell

Childs

(2015) All aboard the pink battle bus? Women voters, women’s issues, candidates and party leaders. Parliamentary Affairs 68(1): 206–223.

Campbell

Lovenduski

(2005) Winning women’s votes? The incremental track to equality. Parliamentary Affairs 58(4): 837–853.

Carroll

(2006) Voting choices: meet you at the gender gap. In: Carroll

Fox

(eds) Gender and Elections: Shaping the Future of American Politics. Cambridge: Cambridge University Press, pp. 74–96.

Carroll

(2008) Security moms and presidential politics: women voters in the 2004 election. In: Whitaker

(ed.) Voting the Gender Gap. Illinois: University of Illinois Press, pp. 75–90.

Carroll

(2010) Voting choices: the politics of the gender gap. In: Carroll

Fox

(eds) Gender and Elections: Shaping the Future of American Politics. Cambridge: Cambridge University Press, pp. 117–143.

10.

Carroll

(2014) Voting choices: how and why the gender gap matters. In: Carroll

Fox

(eds) Gender and Elections: Shaping the Future of American Politics. Cambridge: Cambridge University Press, pp. 119–145.

11.

Carroll

(1999) The disempowerment of the gender gap: soccer moms and the 1996 elections. PS: Political Science and Politics 32(1): 7–11.

12.

Celis

Childs

(2011) The substantive representation of women: what to do with conservative claims? Political Studies 60(1): 213–225.

13.

Celis

Childs

Kantola

. (2014) Constituting women’s interests through representative claims. Politics and Gender 10(2): 149–174.

14.

Celis

Childs

Krook

. (2008) Rethinking women’s substantive representation. Representation 44(2): 99–110.

15.

Childs

(2005) Feminising British politics: sex and gender in the 2005 general election. In: Geddes

Tonge

(eds) Britain Decides: The UK General Election 2005. Basingstoke: Palgrave Macmillan, pp. 155–167.

16.

Childs

(2007) Representation. In: Bryson

Blakeley

(eds) The Impact of Feminism on Political Concepts and Debates. Manchester: Manchester University Press.

17.

Childs

(2008) Women and British Party Politics: Descriptive, Substantive and Symbolic Representation. London: Routledge, pp. 73–91.

18.

Childs

(2015) Feminizing political parties: what stops things’ happening in parties? In: Gender Inequality and Power Commission. LSE. Available at: http://www.lse.ac.uk/genderInstitute/research/commission/home.aspx.

19.

Childs

Webb

(2012) Sex, Gender and the Conservative Party. London: Palgrave Macmillan.

20.

Dionne

Jr (1996) Clinton swipes the GOP’s lyrics; the democrat as liberal republican. The Washington Post, 21 de julio de 1996, C1.

21.

Dolan

(1998) Voting for women in the year of the woman. American Journal of Political Science 42(1): 272–293.

22.

Dovi

(2002) Preferable descriptive representatives: will just any woman, black or latino do? American Political Science Review 96(4): 729–743.

23.

Dovi

(2007) Theorizing women’s representation in the United States. Politics and Gender 3(3): 297–319.

24.

Duverger

(1955) The Political Role of Women. Paris: UNESCO.

25.

Eagle

Lovenduski

(1998) High Time or High Tide for Labour Women? London: Fabian Society.

26.

Frankovic

(1982) Sex and politics: new alignments. Old Issues. P.S 15: 439–448.

27.

Frankovic

(1988) The Ferraro factor: the women’s movement, the polls and the press. In: Mueller

(ed.) The Politics of the Gender Gap: The Social Construction of Political Influence. Newbury Park: Sage, pp. 82–101.

28.

Freeman

(1987) Who you know versus who you represent: feminist influence in the Democratic and Republican parties. In: Katzenstein

Mueller

(eds) The Women’s Movement of the United States and Western Europe: Feminist Consciousness, Political Opportunity and Public Policy. Philadelphia: Temple University Press, pp. 215–244.

29.

Gillion

Ladd

Meredith

(2013) Education, party polarization and the origins of the gender gap in U.S. Party Identification. APSA 2013 Annual Meeting Paper; American Political Science Association 2013 Annual Meeting. Available at SSRN: http://ssrn.com/abstract=2301103.

30.

Hewitt

Mattinson

(1987) Women’s Votes: The Keys to Winning. London: The Fabian Society.

31.

Inglehart

Norris

(2000) The developmental theory of the gender gap: women and men’s voting behaviour in global perspective. International Political Science Review 21(4): 441–462.

32.

Jamieson

Falk

Sherr

(1999) The enthymeme gap in the 1996 presidential campaign. PS: Political Science and Politics 32(1): 13–16.

33.

Jamieson

Waldman

(2004) The Press Effect: Politicians, Journalists, and the Stories that Shape the Political World. Oxford: Oxford University Press.

34.

Kaufmann

Petrocik

(1997) The revenge of the soccer moms? Gender as a party cleavage in American politics. Paper presented at the American Political Science Association, 28–31 August 1997, Washington, USA.

35.

Kittlison

(2006) Challenging Parties, Changing Parliaments. Columbus: Ohio State University Press.

36.

Lipset

Rokkan

(1967) Party Systems and Voter Alignments. New York: Free Press.

37.

Lovenduski

(1997) Gender politics: a breakthrough for women? Parliamentary Affairs 50(4): 708–719.

38.

Lovenduski

(2001) Women and politics: minority representation or critical mass? Parliamentary Affairs 54(4): 743–758.

39.

Lovenduski

(2005) Feminizing Politics. Cambridge: Polity Press.

40.

Mansbridge

(2003) Rethinking representation. The American Political Science Review 97(4): 515–528.

41.

Mueller

(1988) The empowerment of women: polling and the women’s voting bloc. In: Mueller

(ed.) The Politics of the Gender Gap: The Social Construction of Political Influence. Newbury Park: Sage, pp. 16–36.

42.

Nie

Verba

Petrocik

(1979) The Changing American Voter. Cambridge: Harvard University Press.

43.

Norris

(2001) The gender gap: old challenges, new approaches. In: Carroll

(ed.) Women and American Politics: Agenda Setting for the 21st Century. Oxford: Oxford University Press, pp. 146–172.

44.

Norris

Carroll

(1997) The dynamics of the news framing process: from reagan’s gender gap to clinton’s soccer moms. Paper presented at the Annual Meeting of the Southern Political Science Association. >Norfolk Virginia, 5–8 November 1997. Norfolk, USA.

45.

Norris

Lovenduski

(1993) Gender and party politics in Britain In: Lovenduski

Norris

(eds) Gender and Party Politics. London: Sage, pp. 35–59.

46.

Phillips

(1995) The Politics of Presence. Oxford: Oxford University Press.

47.

Russell

(2005) Building New Labour. Basingstoke: Palgrave.

48.

Rymph

(2005) Republican Women: Feminism and Conservatism from Suffrage through the Rise of the New Right. Chapel Hill: The University of North Carolina Press

49.

Saward

2010. The Representative Claim: Oxford University Press.

50.

Schwindt-Bayer

Mishler

(2005) An integrated model of women’s representation. The Journal of Politics 67(2): 407–428.

51.

Seltzer

Newman

Leighton

(1997) Sex as a Political Variable: Women as Candidates and Voters in U.S. Elections. Boulder: Lynne Rienner Publishers, Inc.

52.

Sigel

(1999) Gender and voting behavior in the 1996 presidential election: an introduction. PS: Political Science and Politics 32(1): 5–6.

53.

Tingsten

(1937) Political Behavior. Totowa: Bedminster.

54.

Wängnerud

(2000) Testing the politics of presence: women’s representation in the Swedish riksdag. Scandinavian Political Studies 23(1): 67–91.

55.

Young

(2000) Feminists and Party Politics. Vancouver: University of British Columbia Press.