The Grit in the Oyster? Women’s Parliamentary Organizations and the Substantive Representation of Women

Abstract

This article addresses a foundational question of political representation: how do representatives act for those they represent? In a shift away from analyses of individual representatives’ attitudes and behaviour, we identify Women’s Parliamentary Organizations as potential critical sites and critical actors for women’s substantive representation. Offering one of the most in-depth studies to date, our illustrative case is the long-standing UK Parliamentary Labour Party’s Women’s Committee. With a unique data set, and using both quantitative and qualitative methods, we systematically examine the Parliamentary Labour Party’s Women’s Committee efforts to substantively represent women over more than a decade. We find that the Committee sustains its focus on a small number of women’s issues and interacts with party leadership to advance women’s interests in a feminist direction. Our findings capture processes of political change, a frequently under-explored stage in studies of substantive representation. We close by identifying the potential for comparative research in this area.

Keywords

women’s substantive representation critical mass critical actors political change gender and politics feminizing women’s parliamentary organizations

Introduction

The simple contention that substantive representation flows from descriptive representation has underpinned much gender and politics research over the past two decades. Here, we rethink this classic question by shifting attention away from the behaviour of individual women legislators to analyse Women’s Parliamentary Organizations (WPOs), an umbrella term for various types of women’s committees, caucuses and more informal groups (Celis et al., 2016; Piscopo, 2014; Sawer and Turner, 2016). We contend that WPOs constitute a ‘missing link’ that can bring the relationship between women’s descriptive and substantive representation into better focus (Harder, 2017).

Key questions relating to substantive representation are revisited via a case study of the UK Parliamentary Labour Party’s Women’s Committee (WPLP). Specifically, we systematically identify (1) the actors of women’s substantive representation (which women legislators are members of, and active in, the WPO), (2) the content of women’s substantive representation (how the WPLP defines women’s issues and women’s interests) and crucially (3) the processes by which the group seeks to act for women (how and upon whom the WPLP seeks to have a re-gendering effect). Leveraging original qualitative and quantitative data we are able to demonstrate the existence of a set of women MPs who over time constitute the WPLP’s core membership and fortify the Committee’s work, the WPLP’s ongoing focus on a small number of women’s issues over more than a decade and the Committee’s capacity to interact directly with and – in their view – hold to account the Labour party leadership vis-a-vis what the Committee defines as women’s interests. Our analysis suggests that the WPLP is exemplary of the way in which a self-identified feminist organization can engender political change, even in a highly constrained masculinized context.¹

We begin with a brief review of established theoretical claims linking descriptive and substantive representation, along with a summary of associated criticisms. We then justify our turn to parliamentary organizations ‘for’ women, contending that these have neither in general, nor in the case of the WPLP in particular, been subjected to systematic empirical analysis in respect of substantive representation, with the exception of Harder’s (2017) recent study of a Danish committee. Rejecting a categorical distinction between women’s legislative committees and women’s caucuses, we outline a set of generalizable research questions through which WPOs can be empirically analysed. Our original and unique data and mixed-method methodological approach are then outlined before the presentation of our empirical findings, which are organized to speak of the substantive representation literature’s ‘agreed’ research framework (Childs and Lovenduski, 2013). Overall, we claim that it makes most sense to see WPOs as both sites of, and potential critical actors in, women’s substantive representation within masculinized legislatures. We close with a reflection on how our approach could be exported to comparative or other single-country case studies.

Rehearsing Women’s Substantive Representation

In the ‘politics of presence’ literature, a re-gendering of the content of politics is said to be one likely, albeit not guaranteed, consequence of the changed composition of our elected institutions (Manbsridge, 1999; Phillips, 1995; Williams, 1998). The standard account goes like this: women representatives act for women because they are feminist or at least gender conscious, act in a feminist direction, but do so in institutions that are largely gender-insensitive (Celis and Childs, 2018). A plethora of global empirical research finds much that is positive in this purported relationship (see Childs and Lovenduski, 2013), even as it reveals that processes and outcomes are more complex, contingent and contested than this optimistic account suggests (Celis et al., 2008; Weldon, 2014).

In simple terms, critical mass theory contends that substantive representation will come about as a result of there being a ‘critical mass’ of women in a legislature. Only as their numbers increase – the pivotal figure usually taken to be 30% (Dahlerup, 1988) – will women be able to work more effectively together to promote women-friendly policy change and to influence their male colleagues to accept and approve legislation promoting women’s concerns (Childs and Krook, 2009).

In recent years, this approach has been criticized for a naivety deriving from the concept’s heritage in physics (Childs and Krook, 2009). Unlike science there is no magic in numbers in politics (Beckwith, 2007). The likelihood of women representatives ‘acting for women’ and delivering women’s substantive representation is mediated by a myriad of factors, including their newness, party identity and institutional marginalization (see Childs and Lovenduski, 2013). The political contexts within which women act, not least the gendered nature of parliaments, are less passive backdrops and are instead more constitutive of, women’s substantive representation. Critical mass theory also suggests a theoretically troubling essentialism that assumes women are all the same and stands accused of privileging a universal and feminist definition of women’s interests (Celis and Childs, 2012). Nor can it account for acts for women undertaken by male representatives (Celis et al., 2014), for its claims rest on increases in the numbers and percentages of women. The preferable concept of critical actors (Childs and Krook, 2009) leaves open the identity of those who act for women, defining them in terms of what they do rather than who they are. Male or female, these representatives initiate policy proposals on their own and often – but not necessarily – embolden others to take steps to promote policies for women, regardless of the number of female representatives present in a particular institution (Childs and Krook, 2009: 734). The tendency to focus on the actions of individual elected women representatives, albeit at the aggregate level, has also been called into question. This critique comes from scholarship on women’s movements (Beckwith, 2013; Weldon, 2002), gender mainstreaming and femocrats (McBride and Mazur 2012), women and executives (Annesley and Gains, 2010; Gains and Lowndes, 2014) and non-elected actors within processes of representation (Saward, 2010). Based on such studies the actors of women’s substantive representation are frequently agreed to be multiple and collective, acting within and outside legislatures, and to be both elected and non-elected.

In this context, contemporary scholars of women’s substantive representation are guided to answer eight, linked questions: (1) Why should women be represented? (2) Who are the representatives of women? (3) Which women are represented? (4) Where does the representation occur? (5) How is the representation done? (6) When does it take place? (7) To whom are representatives accountable? (8) How effective is the (claimed) representation? (Childs and Lovenduski, 2013; citing Celis et al., 2008; Dovi, 2007, 2010). This framework informs our analysis as we turn now to consideration of women’s parliamentary organizations as potential sites of, and actors in, women’s substantive representation.

WPOs and Women’s Substantive Representation

WPOs refer to groups also known as women’s parliamentary bodies or ‘women specific legislative initiatives’ (Piscopo, 2014). Two types are usually identified in the practitioner and academic literature: (1) legislative committees – formal groupings that play regular legislative or policy roles addressing women or gender and (2) women’s caucuses – groups of women legislators acting together across party lines (Piscopo, 2014). These are not, however, mutually exclusive: a legislature may have one or both, or even something that matches neither definition, even as it is recognizable or self-identifies as a WPO (Celis et al., 2016).²

The establishment of a WPO tends to follow an influx of women representatives into a legislature, and their presence has become a core criterion when identifying ‘Gender Sensitive Parliaments’ (Sawer and Turner, 2016: 765; www.ipu.org, Childs, 2016). Their creation can be a study of institutional re-gendering, but our enquiry here is narrower, focusing less on how WPOs come about, and more about their role in ‘acting for’ women. As Mette Marie Staehr Harder (2017) puts it (emphasis added): WPOs constitute ‘new arenas for problems, solutions and choice opportunities for actors devoted to this type of representation’ (p. 437). To this we add, WPOs have the potential to constitute themselves as actors in the process of women’s substantive representation when they seek to act collectively.³

We do not see it as necessary to accept the above distinction between women’s committees and caucuses when studying women’s substantive representation (Mitchell-Mahoney, 2013; Piscopo, 2014). Even as we acknowledge that although not all women’s caucuses explicitly include, or should include, a policy or legislative dimension, most do (www.ipu.org; Mitchell-Mahoney, 2013; Oliver 2005).⁴ And in this, they are seeking the substantive representation of women (Piscopo, 2014). Moreover, many caucuses are, like committees, formal, with written rules, well-defined structures, clear membership, ‘public allegiances identifiable’ to members and non-members and official sanctions (Piscopo, 2014: 7). Reasons for participation might very well be the same too (cf Piscopo, 2014: 15). Nor do we assume a priori that committees are inherently more effective than caucuses at delivering substantive representation. This is an empirical question, and likely to reflect criteria other than their formal designation (see Piscopo, 2014). Harder’s (2017) analysis of the Danish Committee on Gender Equality establishes, for example, that even while the Committee did not seek to ‘add’ gender to legislation (Holli and Harder, 2016) nor increase ‘parliamentary specialization’, it was able to ensure ‘parliamentary control’ of the executive, and ‘interaction between parliament and civil society’ (Harder, 2017: 453, 441–442).⁵ Alongside this, we note Mitchell-Mahoney’s (2013) observation that ‘women who meet with other women are slightly more likely to work on women’s issues bills than those who do not’ (pp. 9–10). The IPU also suggest that caucuses are ‘particularly effective in changing legislation and policies from a gender perspective and raising awareness about gender equality’. Moreover, Sue Carroll’s (2001) study of US state legislatures, designed to examine the impact of critical mass, established a relationship between the presence of a women’s legislative caucus and the passage of legislation dealing with women, children and the family.

We define a WPO in an inclusive fashion: a regularized but not necessarily formal association of legislators formed to sustain women’s presence in the political institution, and/or to engender women’s representation, descriptive, substantive, and symbolic. We outline some of the different characteristics, forms, resources and activities that may help identify the ways in which, and the extent to which WPO might be positioned to act for women in Table 1.

Table 1.

Features of Parliamentary Women’s Organizations.

Features	Manifestations
Remit	• Increasing descriptive representation of women in party • Increasing descriptive representation of women in the legislature • Influencing government policy • Influencing party policy platform
Ideology	• Self-defined/identified as feminist • Gendered • Feminist (as matched to demands of women’s movement)
Membership	• Bipartisan versus partisan • Women-only versus men and women • Lower House only versus both Houses
Activities	• Regular meetings for members • Hearings for/with ministers • Hearings for/with non-member MPs • Hearings for civil society • Policy development • Legislative initiation (proactive) • Legislative response (reactive) • Intra- and inter-party mobilization • Social events • Fundraising for recruitment of women candidates
Civil society links	• Outward-facing dimension • Formal versus informal links
Rules	• Formal versus informal • Minimal versus extensive • Presence/absence of sanctions
Resources	• Financial versus in-kind • High or low (relatively to other Parliamentary caucuses/bodies) • Funding source: self-funded by members versus parliamentary versus extra-parliamentary civil society versus extra-parliamentary parties
Degree of institutionalization	• Established (and protected) by institutional rules; degree of immovability; integration within larger institution • Professionalized versus amateur body – for example, staffing arrangements • Frequency of meetings • Degree of formalization of procedures, records and accounts • Stature/standing

Source: drawing on Piscopo, 2014; Celis et al., 2014

Drawing on emergent research, Table 2 sets out a set of core research questions for the study of WPOs. These permit study of their role in the oft-hidden processes that link women’s descriptive and substantive representation,⁶ and in doing so, address the aforementioned ‘8 question’ substantive representation research framework.⁷

Table 2.

Research Questions for Studying Women’s Substantive Representation via WPOs.

1. Was the WPO established to (a) support the generalized capacity of women members? (reactive, apparently non-policy oriented) and/or (b) to voice a gendered/feminist agenda (proactive, explicitly policy-oriented)?
2. How does the group conceive/define ‘women’s interests’ vis-a-vis feminism, and is this (a) the product of its members (internal) or (b) the product of extra-parliamentary groups (co-production or externally produced and subsequently adopted)?
3. How is the WPO membership determined? Is it inclusive or exclusive in terms of sex and party?
4. What activities does it undertake? And how does it engage with those in positions of legislative and executive influence?
5. How extensive is the organization’s capacity and professionalization (level of resources, formalized rules, degrees of institutionalization)
6. How do members sustain their organization?

At the outset, it is useful to determine how a WPO conceives of itself fitting with, and acting within, a legislature. In other words, how is the WPO alert to the masculinized tendencies of the institution in which it sits? Is it seeking to support women representatives as they act as representatives (in the face of institutional sexism)? And/or is the WPO seeking to articulate a particularly gendered issue agenda that it perceives to be otherwise missing? Of course, we should also acknowledge that both a WPO’s motive and focus may change over time.

Membership speaks of the politics of women-only spaces, intra-party cohesion, and inter-party competition. In gender unequal parliaments, the appeal of women only-spaces may seem self-evident to many women legislators (Childs, 2004). These undoubtedly permit the generation of group perspectives (Sawer and Turner, 2016: 767), can engender ‘collaborative relationships’ between women (Barnes, 2016: 48; Childs, 2013) and enhance women’s participation in the policymaking process. But such a choice risks attendant marginalization, reduced effectiveness and can limit who has the opportunity to serve as a critical actor (more of which later).⁸

Distinctions between members relating to participation levels are necessary to more precisely document the work of the WPO; whether some members ‘do’ more than others. There are also issues of sustainability. For example, are women’s regular meetings over dinner merely social, or might they play important roles in creating and reinforcing the bonds that underpin actions more directly related to the women’s substantive representation and the ongoing health of the WPO?⁹

How does the WPO decide upon its policy agenda (where it has one)? This speaks once again of membership, but also of political ideology and democracy, and of which women are substantively represented by the WPO. Cross-party women’s membership might look most attractive: increasing the overall size of a WPO brings symbolic gains, potentially making the WPO the recognized voice of ‘all’ women. Yet, this may give rise to a narrowing in the range of issues acted upon, given the likely need for partisan consensus. It might also render the WPO less effective, substantively speaking, as it prioritizes issues that are not necessarily the most salient but rather those in which the distance of partisan difference is most easily overcome. In contrast, where membership is party specific, there is likely to be greater internal cohesion regarding both what constitutes women’s issues and in respect of what is agreed to be ‘in the interests of women’.

We might also expect WPOs to adopt different modes of activity, depending on their form (e.g. intra- or cross-party, women only or mixed). But WPO action is likely linked also to questions of resources and capacity and legitimacy and power, not least the relationship of the WPO to the party (or parties) of government. We might hypothesize a well-resourced and highly institutionalized women’s organization to be better positioned to deliver substantive representation than one that is poorly funded, relies on the capacity and political capital of individuals and which finds itself marginalized within the institution. Yet, Petra Ahrens’ (2016) analysis of the European Women’s Rights and Gender Equality Committee (FEMM) shows how members negotiated its status, particularly its voluntary membership, to their advantage. Arguably WPOs, regardless of whether they have a formal legislative or scrutiny remit (Holli and Harder, 2016), will need to work with and through those who occupy positions of power within masculinized legislatures. This in turn, raises substantive and theoretical questions about who gets to be regarded as the ‘critical actors’ for women. Without examining the groundwork, or even ground-softening, put in by a WPO, our understanding of how substantive representation comes about is much diminished.

We might also anticipate a WPO acting differently – presenting a different ‘face’ – depending on the actors or audience it is addressing, internally as well as externally, if, indeed, it is addressing external audiences. This begs questions of the WPO’s relationship with those they claim to represent: a WPO open to women beyond the legislature might be considered more responsive and accountable, and hence feminist (Celis et al., 2016; Sawer, 2015). Finally, in a more general sense, a WPO’s strategy for pursuing women’s substantive representation is likely be conditioned by wider political contexts: the health of the economy, for example, or other kinds of fundamental shifts in the lived experience of women in society.

Data and Case Study

The WPLP is a feminist, women-only parliamentary body. When it was established in 1982, its founder Harriet Harman MP pointedly sought both the election of more Labour women MPs and the prioritization of women’s issues and feminist perspectives as part of the party’s policy agenda (Lovenduski, 2005). Previous mainstream and gender research has had little to say about the WPLP. Interviews with Labour women MPs in the 1990s and 2000s suggested that it constituted an important site for women’s substantive representation and work on new Labour women MPs’ friendships suggested this was part of the story of women’s participation in the WPLP (Childs, 2004, 2013). Until now, these claims have not been subject to further empirical analysis.

We draw on three original sources of data and, like Harder (2017), we see the value in combining both qualitative and quantitative research methods.¹⁰ First, extensive qualitative interview data comprising over 40 semi-structured interviews conducted between 2013 and 2015 with individuals associated with the WPLP. Thirteen held executive office, 20 held government office and the rest were ordinary members, including ‘good’, ‘fair’ and ‘poor’ attendees.¹¹ Most interviews lasted around 45 minutes, were recorded at Westminster (2013–2014) and fully transcribed. 2015 interviews included a few ministers – male and female – identified in the first round of interviews as particularly responsive to the WPLP.

Second, an individual-level dataset registering members’ attendance patterns generated from the minutes of WPLP meetings between 2001 and 2015. Attendance is coded on a meeting-by-meeting basis for each person. This permits the computation of overall patterns of attendance as well as limited analysis of predictors of attendance.

Third, a text corpus generated from the minutes of WPLP meetings between 2001 and 2015, though owing to extensive missing or limited data from both 2001 and 2015, we only analyse data from the years 2002–2014 inclusive.¹² Above and beyond facilitating the construction of the individual-level dataset described above, the minutes permit further disaggregation of attendance by internal and external attendees, the identification of WPLP issue priorities, linkages within and outside of the WPLP and a general sense of WPLP activities over the period.

Patterns of Attendance

The WPLP is open to all Labour women MPs and Peers, some 224 potential attendees in the period under study. Reviewing our data, attendance was never a majority activity among Labour women MPs and Peers, numbers of which ranged from 81 to 99 and from 52 to 66 throughout the period, respectively, as shown in Figure 1.¹³ Overall attendance ranged from a high of 44 (June, 2005) to a low of 2 (July, 2012), variation which might mirror the changes in the overall number of possible members shown in Figure 1.¹⁴ There were 180 meetings during this period, three of which we were unable to determine attendance numbers for.¹⁵ Outside of these cases the average number of attendees at meetings (2002–2014) was 15. Attendance varies only slightly based on the presence of external visiting speakers from the Labour frontbench (a mean of 16 attendees compared to 14).

Figure 1.

Women MPs and Peers from the Labour Party 2001–2015.

The question of whose attendance underpins the WPLP can be examined by analysing the raw attendance count data. Looking at this for each entry cohort of MPs, MPs elected prior to 1997 (n = 27) attended an average of 20 meetings, MPs elected in 1997 (n = 61) an average of 23 meetings, 2001 (n = 4) an average of 45 meetings, 2005 (n = 27) 21 meetings, and 2010 (n = 27) 9 meetings. These data suggest that, during our period of study, all cohorts played a relatively equal role, barring the outlier of 2001, which is owing to the uncommonly high rate of attendance of one MP who entered the Commons at that election.

Turning to individual patterns of attendance, for each MP or Peer we calculate the number of meetings they attended as a percentage of those that they were eligible to attend. The number of eligible meetings varies by individual, based on their parliamentary tenure and dates of entry and exit. As such, this percentage figure offers a more accurate measure for comparing attendance patterns than a raw individual count. Figure 2 presents a histogram of the distribution of attendance patterns using this percentage figure. The mean percentage of eligible meetings attended by MPs or Peers is 17%, or just under one out of every five. This ranges from members who attended zero eligible meetings to those who attended 77%, or almost four out of every five.

Figure 2.

Histogram of Distribution of MP and Peer Attendance at WPLP Meetings 2002–2015.

There is some variation in attendance dependent on factors including entry cohort, which House the member sits in and, in the case of MPs, whether they were selected and elected via an all-woman shortlist (AWS). MPs elected for the first time before 1997 (n = 27) attended an average of 18% of eligible meetings, while those MPs elected in 1997 (n = 61) went to an average of 22% of eligible meetings.¹⁶ Looking at the other two cohorts with significant numbers of new women MPs elected, 2005 and 2010, MPs elected at these elections attended an average of 21% and 16% of eligible meetings, respectively. The four women MPs elected for the first time in 2001 attended an average of 30% of eligible meetings. Suggestive of a small quota effect (Childs and Krook, 2012), women elected via Labour’s AWS party quota (n = 33) attended 22% of eligible meetings on average, this in comparison to the 16% attended by their colleagues elected via open lists or who were appointed to the House of Lords (n = 189).

MPs (n = 147) attended almost twice as many eligible meetings than Peers (n = 75), 20% and 11%, respectively. Attendance by frontbench members comes up against the issue of time pressure although the WPLP members are sympathetic to such constraints and the differences are fairly small: the mean percentage of eligible meetings attended by MPs or Peers who served in either the Cabinet or Shadow Cabinet during the period of study (n = 33) is 16% compared to 17% for those who did not hold such positions. In Appendix C, we present regression models of individual-level variation in WPLP meeting attendance that confirm many of these patterns, although our ability to model individual-level attendance extensively is limited by sample size concerns. In terms of the top 10 attendees (Table 3), we see evidence of membership of the Executive Committee (our knowledge of which is detailed in Appendix B) driving these: Julie Morgan, Fiona Mactaggart, Sheila Gilmore, Barbara Keeley, Joan Ruddock, Lorna Fitzsimons and Barbara Follett all held Executive Committee positions and comprise a significant portion of the top 10 attendees.

Table 3.

Top 10 Attendees by Individual Percentage of Eligible Meetings Attended.

Name	%
1. Helen Jackson MP	77
2. Kate Green MP	76
= Julie Morgan MP
3. Fiona Mactaggart MP	73
4. Sheila Gilmore MP	69
5. Sharon Hodgson MP	67
= Julia Drown MP
6. Barbara Keeley MP	65
7. Valerie Davey MP	62
8. Joan Ruddock MP	59
= Rachel Squire MP
9. Lorna Fitzsimons MP	58
= Baroness Whitaker
10. Barbara Follett MP	56

Member Motivations to ‘Act for Women’

The make-up of the WPLP is best understood in terms of a shared gendered identity and a consciousness of gendered interests.¹⁷ Attendees had been active in other women’s organizations within the party; wanted to ‘promote women in the party’; and were ‘carrying the fire for women’s issues’. There was an overwhelming sense that members were feminists.¹⁸ The most popular configuration of reasons for participation is captured by the following MP:

[it was] pretty automatic for me … [it] reflected my interests … a natural thing to do, to meet together with other women members of the PLP … I also knew other women who were going.

That attendance was simply the obvious or natural thing to do was mentioned by 11 interviewees. This was closely followed by prior interest or activism on women’s issues, and then an invocation of it being akin to some kind of duty to attend – both being mentioned by 10 interviewees. Other MPs stated that they attended with their cohort. The relative consistency in attendance rates across the cohorts included in the study, discussed above, offers support for this claim. Interviewees used a number of adjectives and sentiments to describe the WPLP’s atmosphere, style or tone: safe (5 members), informal (5 members), supportive (3 members), comfortable (3 members) and private or trusting (3 members). The tone of the WPLP was said to be in contrast to the meetings of the full Parliamentary Labour Party.

For its members, the WPLP was also a ‘political coming-together’; a party women’s caucus where the women could identify their collective perspectives and policy agenda. This perception evidently underpins the first role of the WPLP: inserting women’s perspectives, issues and interests onto the Party’s and, given they were in power for much of the time, the Government’s agenda. This goal was itself two-fold: first, it is about the ways in which the WPLP would introduce ‘a woman’s angle’. MPs talk about ‘a feminist approach to politics and policymaking that we have got to get across the whole gamut of policy’. Second, it is about articulating particular women’s issues, ‘things that only women bother about’, and issues that were high in the priority list of women’. The starting point was a shared perception that in the absence of ‘reminding the guys’ – male ministers, young male advisors and the civil service – to feminize their policies, ‘things’ will get neglected. In their words, the group was ‘breaking new ground’, ‘pushing new boundaries’.

Women’s Issues and the Interests of Women

Of interest is whether the WPLP focused on particular women’s issues during the period of study. Beginning with our interview data, we find that members explicitly invoke a number of issues under the rubric of ‘women’s issues’: childcare, violence, pensions, the institution of Parliament itself and caring. All were mentioned multiple times. These issues were said to be ‘bottom up’ – stemming from the ‘greater contact’ women MPs felt they had with women and women’s voluntary groups. That said, three critics of the WPLP suggested these concerns reflected a metropolitan feminism, and or group members’ self-interest.

Building on this qualitative analysis, we leverage the unique text data included in the WPLP minutes and use structural topic models (STM) to computationally estimate the topics discussed by the group.¹⁹ STM offers a more objective purchase than qualitative analysis alone, specifically giving tools to (1) identify the topics of discussion at the meetings over the period of study, (2) consider how topic prevalence varies conditional on time and/or the individuals present at a meeting and (3) see the extent to which individual politicians are associated with given topics. The primary purpose of using STM is to identify the issues the WPLP discuss in meetings – in other words, the ‘topics’ of the aforementioned topic model. To do this, we estimate an STM with a covariate for the year in which the meeting took place, taking into account the fact that we would expect certain topics of discussion to vary in prevalence across the period of study. Based on the initial simple model, which analyses minutes from meetings between 2002 and 2014, four topics of discussion are identified. We compared models with varying numbers of identified topics (see Appendix A) and, based on the diagnostic values for each, settled on four as the optimal number.

We then generated word clouds (top 10 most frequent words for each topic) for the topics in order to explore two of our research questions: what the topics are and who, if anyone, appears to be associated with them.²⁰

As shown in Figure 3, Topic 1 is the procedural topic and is associated with WPLP’s internal workings. Terms relating to meetings, such as ‘held’, ‘record’, ‘contribute’ and ‘raise’ are prevalent. Topic 2 is concerned with violence against women and women’s bodily integrity. Terms such as ‘violence’, ‘equal’, ‘campaign’, ‘consult’ and ‘debate’ indicate the activities of the WPLP in this particular issue area. Topic 3 focuses on issues around care and caring. Terms like ‘care’, ‘childcare’ and ‘older’ are prevalent. Finally, Topic 4 focuses on the question of pensions and the wider gendered nature of the economy. Key terms in this topic include ‘pensions’ (former Chancellor and Prime Minister), ‘Gordon Brown’ and ‘bill’. When reviewing the topics in light of the qualitative findings above, we considered them to be plausible portrayals of the overall activities of the group and the identification of these topics via STM allows us to address our research question regarding the content of substantive representation undertaken by the WPLP.

Figure 3.

Word Clouds for Each Topic.

Of interest following topic identification is the extent to which these topics are consistent in their prevalence across the period of study. In Figure 4, we examine the relative prevalence of each topic in the period 2002–2005 and 2010–2014 and find that although some topics are seen marginally more in one period than the other, there are no statistically significant differences in the prevalence of any topic between the two periods. This suggests that the issue focus of the WPLP was largely consistent throughout the period of study. This is notable given the two periods compared here are quite distinct: between 2002 and 2005, Labour formed a strong second-term government in a time of economic growth whereas between 2010 and 2004, Labour were in opposition following a devastating economic crash and recession. As such, the fact that we do not see significant shifts in topic prevalence across these two time-periods is indicative of a strong and continuing preoccupation with the issues in question on the part of the WPLP.

Figure 4.

Comparison of Topic Prevalence in 2002–2005 and 2010–2014 Meeting Minutes.

Representative Acts

In promoting women’s issues and interests, the WPLP is claimed by many of its members to have successfully re-gendered the Party’s and Government’s agenda. MPs use a range of metaphors and analogous descriptions: the ‘grit in the oyster’, ‘sleeping dragon’, ‘bending the ears’ of Ministers, ‘sharp heels and sharp elbows’ and as a ‘solid, nice glowing core presence’. WPLP members considered that they effected change in a centrifugal fashion, bolstered by a sense of collective identity. For a large number of women MPs, this was also about supportive relationships with women ministers who attended the group.²¹ Not only did this secure some vertical accountability, it was also claimed to empower the ministers. According to one minister:

Without the WPLP … I would have been a minnow, at the bottom of an anonymous department … nothing would have happened…How do you get … [legislation] into the Queen’s speech? … the WPLP committee was a reason – a side within a side (emphasis added).

Nonetheless, a group of women ministers who claimed to attend the WPLP while they were in office did not feel that the group was determining of their ministerial work, even as they felt ‘of the group’ and shared its ends. For Labour’s official Ministers for Women the relationship was said to be more direct. The attendance figures discussed earlier confirm that ministers (and shadow ministers) continued to attend the WPLP at a similar rate to non-ministers.

There are also invited speakers – broadly defined as anyone addressing the group as a whole – at 91% of WPLP meetings. At 60% of meetings, this involves a representative of the Labour frontbench (either Cabinet or Shadow Cabinet), at 25% a member of party staff, usually from Millbank (the party’s headquarters) and at 26% of meetings there is also a further speaker external to both the parliamentary party and party headquarters. Often, this is a representative from a woman’s advocacy group such as the Fawcett Society or Labour Women Councillors Association, or a guest academic or expert speaker.

A relevant question is whether the WPLP spoke about their core issues, identified previously via STM, in a different way when frontbench representatives of the Party were present at meetings. In Figure 5, we analyse the within-topic prevalence of words conditional on whether a frontbench representative was present at a meeting. In the figure, the presence of a frontbench representative is indicated by ‘1’ and the absence of any such figure by a ‘0’. Figure 5 can be interpreted as follows: if a word is strongly to the right, in the section of the figure labelled ‘1’, it can be considered to be a term that is more prevalent within the topic in the minutes of meetings where a frontbench representative was present as a speaker. An example of this would be ‘older’ within Topic 3 in Figure 5. The more words that collect in the centre of the figure, the more we can interpret them as being equally likely to arise within a topic whether a frontbench representative was present or not. As such, topics having fewer words in the centre of the figure can be seen as indicative of greater differentiation in the way the WPLP talked about that topic conditional on the presence of frontbench speakers. In Figure 5, Topic 4 can be seen as an example of this, whereby the terms used to discuss pensions and the gendered nature of the economy differ reasonably markedly depending on the presence of a frontbench speaker. Conversely, Topic 1 seems to be largely discussed in similar terms regardless of this, something that is perhaps to be expected given its general content. Overall, though, Figure 5 offers some evidence that the WPLP shifts the way in which it discusses key issues of interest when frontbench members are present. This is perhaps indicative of a deliberate strategy on the part of members who seek to communicate with these higher profile individuals in the most striking and effective way possible. Returning to the word clouds presented in Figure 3, this might include speaking about pension reform to Gordon Brown in his capacity as either Chancellor or Prime Minister in a different way from how one would discuss this with a fellow backbencher.

Figure 5.

Comparison of Intra-topic Prevalence Conditional on Presence of Frontbench Representative.

What of other ministers formally invited to come before the WPLP? Such invitations are considered central to the group’s actions and the potential to effect substantive representation. Of course, ministers might seek out opportunities to present at the WPLP. Unfortunately, it is not possible to determine the relative prominence of either approach from the data. There was, notwithstanding, a strong sense that invited ministers would almost always attend. The WPLP was regarded by the party (at least as perceived by its members) as the legitimate voice of women in the parliamentary party. It was not a group to ‘piss off’. One member put it more prosaically:

They would sooner be caught running around the tea room with their pants down than they would be caught saying anything wrong … insulting or disparaging about the WPLP … but in the back of their minds – now I am totally putting thoughts into their heads – they are probably thinking, ‘what are they up to?’

A minister concurred: ‘I wouldn’t dare not to go’. Such statements unfortunately cannot tell us whether attendance is for substantive (honourable) or symbolic/tokenistic (dishonourable) reasons. The more engaged male ministers were claimed to be feminist men, younger ones ‘married to younger generations of women’ or ‘feminists’. The interaction was considered ‘top down’ by six interviewees, ‘bottom up dialogue’ by five and ‘questioning Ministers’ by four. The WPLP was, then, far from a passive group – something supported by the evidence in Figure 5 indicating a shift in approach when frontbench speakers were present.

Claims of political change by the WPLP are broad: influence is claimed in the areas of pensions; welfare reform; ‘Surestart’ and childcare; maternity and paternity leave and pay; domestic violence, rape, prostitution and trafficking; sex and relationships; women’s, heath, social care; education; employment; international development; single women; and the economy and the budget. Broader still were claims to have influenced general election manifestos, and the minutes show that Labour party officials and MPs responsible for drawing up the manifesto routinely attended the group. Once again, there was a shared perception that in the absence of women’s integration in the PLP proper, a sufficiently powerful women’s officer, and when ‘time is short’, overwhelmingly male ministers and advisors just ‘forget that there was a female perspective’.

The difficulty, if not impossibility, of fully quantifying the WPLP’s impact is acknowledged by its members. One measure is the perception of the WPLP’s standing. We have already noted the assumption that ministers would not dare to avoid the Committee. On a negative to positive continuum, the most widely held perception amongst the interviewees, albeit in small numbers, is that amongst men and older women at Parliament there are critical views of the WPLP. That said, there is clearly disagreement amongst the women about whether it is accorded legitimacy, faces antipathy from male MPs and Peers and/or whether it is publicly acknowledged but privately problematized, as in the quotation above.²²

Critical Actors

The concept of critical actors was conceived as a counterpoint to the assumptions of critical mass theory that women’s substantive representation would magically occur when sufficient numbers of women are elected to our parliaments (Childs and Krook, 2006, 2009). It appears, however, to have since become something of a ‘catch all’ term; used in ways that suggest that anything any representative does in the implementation of a policy ‘for women’ turns them into a, if not ‘the’ critical actor. We are sceptical of this usage: the original conceptualization was intended to specify particular actors who acted in conditions where women’s issues, perspectives and interests were marginal, and unlikely to be well received (Childs and Krook, 2009). The effort and risk that a particular representative might need to expend was explicitly recognized in the development of the concept. Accordingly, we are sceptical that the ‘final’ individual who ‘flicks the switch’ on a gender policy should be automatically be considered a critical actor. Our approach here ensures we do not fall into the trap of making this assumption, offering more modest conclusions about who might constitute the critical actors within the workings of the WPLP. Strong conclusions about who constitutes the critical actor require a substantial process-tracing of a particular policy outcome.²³

Our text data generated from the WPLP minutes permits us to identify the extent to which named individuals are associated with given topics, albeit with the caveat that identification at this stage is not necessarily associated with future activity, successful activity or indeed anything other than speaking about the issue within WPLP meetings. That said, the combination of data and method that we employ allows us to identify potential critical actors at this stage of the process in a way that avoids having to reverse-engineer accounts of their activities in a post hoc fashion.

In Table 4, we offer a full summary of the topic model that is useful when attempting to identify potential critical actors. Looking at the work on violence against women (Topic 2), we can see the role of two stakeholder organizations – the All Party Parliamentary Group on Violence Against Women, and the Labour Women Councillor’s Association, both of whom liaised with the WPLP in their work. Similarly, on care (Topic 3), we observe that Sheila Gilmore MP played a substantial role in discussions. Finally, on pensions (Topic 4), we see the role played by a range of MPs as well as the then Chancellor and, later, Prime Minister (PM) Gordon Brown and the public opinion pollster Deborah Mattinson. Again, although we are unwilling to identify these individuals as critical actors solely on the basis of these models, we are able to say that they were present, active and influential in these discussions. Intriguingly, they point to different types of actors – individual women MPs, collective organizations of cross-party MPs united around a shared political interest and individuals and organizations linked with Labour – participating and seemingly leading on women’s substantive representation. As noted above, such findings open the door to more focused process-tracing qualitative research guided by statistical findings, not just researcher intuition.

Table 4.

Full Summary of 4-topic STM.

Topic 1 Top Words: • Highest Prob: minist, offic, place, held, contribut, record, select • FREX: report, barbara, held, prime, behalf, record, gould • Lift: afghanistan, prosser, val, behalf, davey, deleg, follett • Score: woman, jackson, aris, wednesday, behalf, respect, heldTopic 2 Top Words: • Highest Prob: violenc, campaign, equal, govern, labour, day, mps • FREX: consult, cut, group, violenc, march, visit, send • Lift: appg, traffick, broadcast, councillor, onlin, aid, sex • Score: traffick, cut, can, safeti, appg, team, happenTopic 3 Top Words: • Highest Prob: will, labour, care, older, polici, make, issu • FREX: older, care, will, age, men, societi, better • Lift: hit, meal, membership, travel, audit, drive, gilmor • Score: older, membership, will, just, can, hit, careTopic 4 Top Words: • Highest Prob: brown, bill, gordon, end, opportun, pension, bns • FREX: gordon, brown, bns, bill, sylvia, kelli, hodg • Lift: gordon, kali, latest, mountford, kelli, while, supplement • Score: latest, gordon, brown, sylvia, moran, kelli, JacksonA topic model with 4 topics, 178 documents and an 813 word dictionary.

Critical Sites

The potential for WPOs to act as sites for the substantive representation of women is clear; for some it is their raison d’etre. That said, this potential is not always realized nor can easily be inferred from their form. Institutionalization may not be as important for the likely effectiveness of WPOs as might be first assumed (Ahrens, 2016). Members of the WPLP agree. As an organization, the WPLP’s informality was considered to be an asset, enabling it to sustain itself over time, not least because, as one member noted, this is why ‘so many women from such a wide range of places in the party’ ‘come along’. Indeed, one former executive member and minister recalled having informal chats with Brown in a swimming pool. That said, if the Group’s reputation was mostly felt to be sufficient without more formal rules, one member recalled how there had been an agreement that documents would be read by the WPLP before publication, but that this had not happened. Another considered that it would be a “sign of success if it automatically got shown (draft policies)” or ‘if somebody said, “you know we are trying to get this policy, can we come and talk to the women’s group”’.

Discussions of the groups’ interactions with Labour’s two prime ministers across the period revealed further how members identified gender as often a marginal concern for both. Neither would have women’s issues at the ‘central core of their thinking’, even if both were ‘respectful’. Brown is regarded as more receptive. Amongst those who discussed this, two critical comments were assigned to both, but Brown received nine positive comments (these bearing on women’s substantive representation), with even one critical comment implying that pressure from the WPLP had influenced him, citing flexible working policies. Brown was said to trust and worry about Harriet Harman, presumably in her role as ‘big sister’ of the WPLP; attending and engaging with the WPLP to mobilize support on particular issues, and in taking them ‘dead serious’. As one member put it: ‘I don’t think it was just, “oh god, it’s those bloody women again, let’s keep them on board”’.

Conclusion

Our contention that existing empirical studies of women’s substantive representation are limited was rooted in a consideration of what was hidden in the everyday assumption that change in politics follows the same laws as physics. The contribution we have made in returning to the classic question of the relationship between women’s descriptive and substantive representation lies in our focus on collective rather than individual acts of representation, specifically the role played by WPOs in acting for women. The study of WPOs remains nascent in the gender and politics sub-field, and analyses of their role in respect of substantive representation are few (see Ahrens, 2016; Harder, 2017). In this article, we have made what we consider to be a significant step forward in this regard.

Our overall research question was simple: ‘what was going on’ as one WPO sought to act for women over a lengthy period of time? The following statement by one of the WPLP members encapsulates our considered conclusion: ‘We have laid siege to this institution, but we are not yet in it’. Nor, we would add, ‘of’ it. The UK Parliament remains unquestionably gender unequal (Childs, 2016). Those seeking to act for women will find themselves facing and negotiating much that recent feminist institutionalist analyses have documented masculinized political institutions (Mackay, 2010, 2014; Mackay and Waylen, 2009; Waylen, 2010; Childs, 2016). As they remind us, the institutional tendency is resistance to change. But political actors can and do seek to exploit institutional tensions. While WPOs should not be thought of as a ‘magic button’ (Harder, 2017: 453), our research illuminates some of these ways in which the WPLP – as an enactor of political change – sought to re-gender the party’s and government’s political agenda.²⁴

Our empirical observations enable us to develop an explanatory account of the acts and processes that engender substantive representation by WPOs nested within masculinized legislatures. We also offer a systematic account of the actions of a WPO that is explicitly seeking to act in what it considers a feminist fashion. In this, we have moreover sought to speak as directly as we can to the eight, linked questions that frame contemporary studies of women’s substantive representation (Childs and Lovenduski, 2013). We contend on the basis of our data that the WPLP was both a critical site and a critical actor in the ‘ripening’ of the party and Labour government (Mansbridge, 2016).

With our unique quantitative and qualitative data, we have been able to better answer why some of Labour’s women MPs participated in the WPO and why many did so over many years. In the face of gender inequality, WPLP members wanted to act ‘for women’ by influencing the party and the government, and did so because they identified as feminists. In terms of the content of women’s substantive representation – ‘what is in the interests of women’ – the qualitative and quantitative data reinforce one other. There was a broad and consistent set of women’s issues over time that constituted the group’s main agenda for change. This was defined in an unapologetic feminist direction by Labour’s women. From this agenda of women’s issues, and from how the WPLP conceived of what is in the interests of women, it is possible, albeit indirectly, to discern ‘which women’ the WPLP considered that they were acting for.

The data additionally provide important ways of understanding how the WPLP was experienced by members – as an individual and collective resource – and how it was perceived by non-member legislators and ministers. In addition to offering resistance to, and the means by which to better negotiate and challenge, masculinized practices and culture, the WPLP is acclaimed as an important site for instigating processes of SRW. It is where women’s issues are discussed and what is in the interests of women constituted. Claims to its effectiveness were held to be (positively) determined by the WPLP’s reputation as the legitimate ‘voice’ of women in the party. This is an insight objectively recorded in the listing of senior party figures attending WPLP meetings, and yet is also something that could only be fully captured in concert with qualitative methods. The WPLP’s reputation is evidently much greater than the overall attendance of WPLP meetings suggests, as documented by our quantitative data. Neither was there undue concern regarding the group’s institutionalization within the Parliamentary party, a position that would be missed if one merely looked at the WPLP’s formal status, capacity and powers.

Our approach could be exported to the study of WPOs either comparatively or in other countries, with likely variation across the features outlined in Table 1 within different political systems. We might expect WPOs to adopt distinct strategies, and to strike a different balance between being either a site of SRW or an actor in it, depending on the exigencies of a given political structure. We also might anticipate the location of executive power, and therefore policymaking clout, within a polity to affect how a WPO operated in pursuit of its goals. Similarly, the extent to which gender equality is a political issue of prominent concern might affect the ways in which the demands of a WPO are framed, possibly as either more or less explicitly feminist, or indeed couched in the language of more mainstream political debate, depending on how members anticipate their wider reception. Overall, exploring how WPOs function in differing institutional and extra-institutional contexts should be the basis of an exciting and important research agenda that will both enhance and challenge existing knowledge on gender, politics and representation. Here, we have made an initial contribution to this endeavour.

Supplemental Material

PSX793080_Supplemental_Material – Supplemental material for The Grit in the Oyster? Women’s Parliamentary Organizations and the Substantive Representation of Women

Supplemental material, PSX793080_Supplemental_Material for The Grit in the Oyster? Women’s Parliamentary Organizations and the Substantive Representation of Women by Peter Allen and Sarah Childs in Political Studies

Footnotes

Acknowledgements

Thanks to the British Academy Small Grants Scheme for funding this research. Many thanks to the members and staff of the Parliamentary Labour Party Women’s Committee for providing access to the materials used in the preparation of this article. Thanks especially to interviewees for giving up their time. We would also like to thank audiences at the 2014 and 2015 American Political Science Association Annual Meetings, the 2015 and 2017 European Conferences on Politics and Gender, the 2016 EPOP Conference, the Study of Parliament Group, as well as the four anonymous Political Studies reviewers for their feedback and suggestions.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors received support from the British Academy Small Grants Scheme.

Supplementary information

Additional supplementary information may be found with the online version of this article.

Appendix A – Model description and comparison statistics generated in the STM package in R (Roberts et al. 2015)

A1 – Description of modelling approach

Figure A2 – Semantic coherence and exclusivity by number of topics

Figure A3 – Diagnostic values by number of topics

Appendix B – PLPWC Leadership 2002-2014 based on AGM Minutes

Appendix C – Modelling individual-level variation in attendance at WPLP meetings

C1 – Model Choice and Specification

Table C2 – Zero-inflated negative binomial regression models of raw individual MP and Peer attendance counts at WPLP meetings

Notes

Author Biographies

Peter Allen is Reader in Comparative Politics and a Deputy Director of the Institute for Mathematical Innovation at the University of Bath. His book, The Political Class: Why It Matters Who Our Politicians Are, was published by Oxford University Press in April 2018.

Sarah Childs is Professor of Politics and Gender at Birkbeck, University of London. She is the author and editor of multiple books including Deeds and Words (2015) and Gender, Conservatism and Political Representation (2015). She is also one of the founding co-editors of the European Journal of Politics and Gender.

References

Ahrens

(2016) The Committee on Women’s Rights and Gender Equality in the European Parliament. Parliamentary Affairs 69 (4): 778–793.

Annesley

Gains

(2010) The Core Executive: Gender, Power and Change. Political Studies 58 (5): 909–929.

Barnes

(2016) Gendering Legislative Behaviour. Cambridge: Cambridge University Press.

Beckwith

(2007) Numbers and Newness: The Descriptive and Substantive Representation of Women. Canadian Journal of Political Science 40 (1): 27–49.

Beckwith

(2013) The Comparative Study of Women’s Movements. In: Waylen

Celis

Kantola

et al . (eds) The Oxford Handbook on Gender and Politics. Oxford: Oxford University Press, pp. 411–436.

Carroll

(2001) Representing Women: Women State Legislators as Agents of Policy-Related Change. In: Carroll

(ed.) The Impact of Women in Public Office. Bloomington: Indiana University Press, pp.3–21.

Celis

Childs

(2012) The Substantive Representation of Women. Political Studies 60 (1): 213–225.

Celis

Childs

(2018) Good Representatives and Good Representation. Political Science & Politics 51 (2): 314–317.

Celis

Childs

Curtin

(2016) Specialized Parliamentary Bodies and the Quality of Women’s Substantive Representation’. Parliamentary Affairs 69 (4): 812–829.

10.

Celis

Childs

Kantola

et al . (2014) Constituting Women’s Interests through Representative Claims. Politics & Gender 10 (2): 149–174.

11.

Celis

Childs

Kantola

et al . (2008) Rethinking Women’s Substantive Representation. Representation 44 (2): 99–110.

12.

Childs

(2004) New Labour’s Women Mps. London: Routledge.

13.

Childs

(2013) Negotiating Gendered Institutions. Politics & Gender 9 (2): 127–151.

14.

Childs

(2016) The Good Parliament. Bristol: Bristol University. Available at: http://www.bristol.ac.uk/media-library/sites/news/2016/july/20%20Jul%20Prof%20Sarah%20Childs%20The%20Good%20Parliament%20report.pdf

15.

Childs

Krook

(2006) Should Feminists Give Up on Critical Mass? A Contingent Yes. Politics & Gender 2 (4): 522–530.

16.

Childs

Krook

(2009) Analyzing Women’s Substantive Representation: From Critical Mass to Critical Actors. Government and Opposition 44 (2): 125–145.

17.

Childs

Krook

(2012) Labels and Mandates in the UK. In: Franceschet

Krook

Piscopo

(eds) The Impact of Gender Quotas. New York: Oxford University Press, pp.89–102.

18.

Childs

Lovenduski

(2013) Representation. In: Waylen

Celis

Kantola

et al . (eds) The Oxford Handbook of Gender and Politics. Oxford: Oxford University Press, pp.489–513.

19.

Childs

Withey

. (2006) A Simple Case of Feminist Cause and Feminist Effect? Reducing the VAT on Sanitary Products in the UK. Parliamentary Affairs 59 (1).

20.

Dahlerup

(1988) From a Small to a Large Minority: Women in Scandinavian Politics. Scandinavian Political Studies 11 (4): 275–298.

21.

Dovi

(2007) The Good Representative. Malden, MA: Blackwell.

22.

Dovi

(2010) Measuring representation: Rethinking the role of exclusion. Paper presented at the American Political Science Association Annual Meeting, Washington, DC, 2–5 September.

23.

Franceschet

(2017) Disentangling Informality and Informal Rules. In: Waylen

(ed.) Gender and Informal Institutions. London: Rowman & Littlefield, pp.115–135.

24.

Gains

Lowndes

(2014) How is Institutional Formation Gendered, and Does it Make a Difference? Politics & Gender 10 (4): 524–548.

25.

Hammond

Mulhollan

Stevens

Jr (1985) Informal Congressional Caucuses and Agenda Setting. Western Political Quarterly 38 (4): 583–605.

26.

Harder

MMS

(2017) Assessing the Impact of Parliamentary Design: The Case of the Danish Committee on Gender Equality. Scandinavian Political Studies 40 (4): 434–456.

27.

Holli

Harder

MMS

(2016) Towards a Dual Approach. Parliamentary Affairs 69 (4): 794–811.

28.

Lovenduski

(2005) Feminizing Politics. Cambridge: Polity Press.

29.

Mackay

(2010) Conclusion. In: Krook

Mackay

(eds) Gender, Politics and Institutions. Basingstoke: Palgrave Macmillan, pp.181–196.

30.

Mackay

(2014) Nested Newness, Institutional Innovation, and the Gendered Limits of Change. Politics & Gender 10 (4): 549–571.

31.

Mackay

Waylen

(2009) Feminist Institutionalism. Politics & Gender 5 (2): 237–237.

32.

Manbsridge

(1999) Should Blacks Represent Blacks and Women Represent Women? The Journal of Politics 61 (3): 628–657.

33.

Mansbridge

(2016, June 9) Honourary Doctorate Lecture. Edinburgh: The University of Edinburgh.

34.

McBride

Mazur

(2012) Gender Machineries Worldwide. World Development Report: World Bank, pp.92.

35.

Mitchell-Mahoney

(2013) Women as Party Members: What Women’s Caucus Creation Tells Us. In: American Political Science Association annual conference.

36.

Oliver

(2005) Women’s Legislative Caucuses. National Conference of State Legislatures. LegisBrief 13 (29): 1–2.

37.

Phillips

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

38.

Piscopo

(2014) IPSA Annual Meeting. Prepared for presentation at the International Political Science Association World Congress Montreal, Canada 19–25 July 2014.

39.

Saward

(2010) The Representative Claim. Oxford: Oxford University Press.

40.

Sawer

(2015) Beyond Numbers. Australasian Parliamentary Review 30: 105–122.

41.

Sawer

Turner

(2016) Specialised Parliamentary Bodies. Parliamentary Affairs 69 (4): 763–777.

42.

Waylen

(2010) Gendered Institutionalist Analysis. In: Krook

Mackay

(eds) Gender, Politics and Institutions. Basingstoke: Palgrave Macmillan, pp.147–162.

43.

Waylen

(2017) Analysing Gender in Informal Institutions. In: Waylen

(ed.) Gender and Informal Institutions. London: Rowman & Littlefield, pp.1–21.

44.

Weldon

(2002) Beyond Bodies. The Journal of Politics 64 (4): 1153–1174.

45.

Weldon

(2014) Using Statistical Methods to Study Institutions. Politics & Gender 10 (4): 661–672.

46.

Williams

(1998) Voice, Trust and Memory. Princeton, NJ: Princeton University Press.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.16 MB