Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Mars casts his ballot: men and the gender gap in Canadian elections Steele, Andrew Morgan 1998

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1999-0096.pdf [ 4.72MB ]
Metadata
JSON: 831-1.0089138.json
JSON-LD: 831-1.0089138-ld.json
RDF/XML (Pretty): 831-1.0089138-rdf.xml
RDF/JSON: 831-1.0089138-rdf.json
Turtle: 831-1.0089138-turtle.txt
N-Triples: 831-1.0089138-rdf-ntriples.txt
Original Record: 831-1.0089138-source.json
Full Text
831-1.0089138-fulltext.txt
Citation
831-1.0089138.ris

Full Text

M A R S CASTS HIS B A L L O T : M E N A N D THE GENDER GAP IN C A N A D I A N ELECTIONS by A N D R E W M O R G A N STEELE B.A. (Hons.), Trent University, 1997 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF M A S T E R OF ARTS in THE F A C U L T Y OF G R A D U A T E STUDIES (Department of Political Science) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A December 1998 © Andrew Steele, 1998 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia Vancouver, Canada Date DE-6 (2/88) Abstract This study argues that previous investigations of the gender gap have concentrated almost exclusively on the behaviour of women voters and have underestimated the electoral significance of men. Employing public opinion surveys and rational choice theory of coalitions, it contends that men's voting behaviour is a key factor in modern elections and that by investigating male voters as people affected by their gender, the gender gap can be better explained. The study finds that the relative importance of the gender gap in Canada may be declining as parties contending to form the government display less gender division in their support, and significant gender differences in the 1997 election are found only in the more extreme parties, like the New Democratic Party and, especially, the Reform Party. Significant gender-related support for the Liberal Party is found to be concentrated in the Trudeau era. The gender gap in Reform Party support is attributed to differences over capitalism, feminism and the use of force. A theoretical model of gender block behaviour is developed using rational choice theory, and the power of the male voting block is demonstrated. Cohesion, elasticity, positioning, size and turnout are identified as important measures of block power, with cohesion and elasticity the most important variable in the gender gap. The gender gap is shown to not be an automatic advantage for women, and that sometimes it works against women's interests. The final chapter discusses the effect of situational and socialisation constraints on attitutudes towards violence, 'masculinized opportunity' and the reactionary backlash against feminism. Table of Contents Abstract Table of Contents List of Tables List of Figures Acknowledgements and Dedication Chapter One Introduction Chapter Two Just What is this Gender Gap Thing? A Short History of Gender in Political Science The Canadian Gender Gap Chapter Three What Lies Under the Surface? The Literature in Review Political Ideology and the Gender Gap Relative Explanatory Power Chapter Four Problems in the Literature Causes of the Focus on Women Chapter Five Voting Blocks and the Gender Gap Reconceiving the Gender Gap Testing the Framework Applying the Framework to Canada Targeting Men Chapter Six Discussion and Conclusions Men and Reform Conclusions Bibliography Appendix One The Refinement of Data from the 1974-1988 CNES Studies Appendix Two The Construction of Attitudinal Indexes - i i i -List of Tables Table 2.1 Table 2.2 Table 2.3 Table 3.1 Table 3.2 Table 3.3 Table 3.4 Table 5.1 Table 5.2 Table 5.3 1980 U.S. Presidential Election Results divided by Gender. Canadian Federal Election Results Divided by Gender 1974-1988 (Refined Data). The Effect of Significant Issue Indexes Regressed on the Gender Gap in Reform Party Support 1997 (Three Variables). The Effect of Significant Issue Indexes Regressed on the Gender Gap in Reform Party Support 1997 (Four Variables). The Effect of Significant Issue Indexes Regressed on the Gender Gap in Reform Party Support 1997 (Five Variables). The Effect of Significant Issue Indexes Regressed on the Gender Gap in Reform Party Support 1997 (Four Variables). Gender Gap in U.S. House Election 1978-1996 (Percentage Voting Democrat) Electoral Cohesiveness - Mean Percentage Vote for the Largest Party in Canadal 974-93 and 1984-93 (Refined Data) Electoral Mobility - Mean Percentage Vote for Gender Dominated Party Groupings 1974-93 (Refined Data) 10 Canadian Federal Election Results Divided by Gender 1993 and 1997 12 Table 5.4 Most Important Issue in 1997 Election Divided by Gender 24 25 29 30 53 56 59 64 -iv-List of Figures Figure 2.1 Gender Gap in Support for the Liberal Party 1974-1997. 14 Figure 2.2 The Gender Gap in Each Canadian Party 1974-1997. 15 Figure 3.1 Graphic Representation of Model Testing Attitudinal Intervening Variables and the Gender Gap in Reform Party Support 23 Figure 4.1 The Gap in Support for Government Spending in the United States (From Deitch) 36 Figure 5.1 Gender Gap in U.S. House Elections 1978-1996 (Percentage Voting Democrat) 54 Figure 5.2 Electoral Cohesion - Mean Percentage Vote for Gender Dominated Party Groupings 1974-93 (Refined Data) 62 -v-Acknowledgements and Dedication While the errors, oversights and omissions of this thesis ^ are mine alone to claim, the credit for its completion and composition must be shared. First and foremost, UBC's Green College and its residents deserve thanks for providing me with food, shelter, warmth and human contact during my writing. In particular, I want to single out Steffan Riddell, Reza Rajabiun, Chris Spearin, Vanessa Timmer, Melanie Bidali, Isabella Varella and Paul Bergbusch for keeping me from working as often as possible. Que sara sara. In the U B C Political Science Department, the principal credit belongs with my supervisor, Richard Johnston. Without his observations this project would never have been possible. Barbara Arneil's advice on matters of gender and politics was outstanding, and for that I am grateful. Thanks to Don Blake for criticism of an early draft and David Green for insight on labour economics. A round of applause for Alison Warner for inspiring so many good ideas, and criticising my poor ones. Kathy Vandergrift was a great help when researching the Reform Party. Finally, I want to acknowledge my family, for their support during the writing of this thesis and for so many years before. Thanks to my parents, Sandra and Tom, my brother Matt, and all my stepparents, grandparents, uncles, aunts and cousins for their encouragement, advice, and the occasional ass-kicking. This thesis is dedicated to the memory of Rob de Jong, scholar, sculler and friend. -vi-Chapter 1 - Introduction: "At twenty a man is a peacock, at thirty a lion, at forty a camel, at fifty a serpent, at sixty a dog, at seventy an ape, at eighty, nothing at all." Baltasar Gracian The study of gender in political science has always been coloured by the normative beliefs of the researchers who undertook its study. At first, women were ignored as a significant group in society and gender itself was disdained as a minor influence on political behaviour. With the advent of second-wave feminism and the mass reappraisal of scholarship that took place in its wake, the study of women gained ground in political science. However, contemporary researchers continue to ignore the effects of gender on men in apolitical context. To simply "add women and stir" was held to be sexing politics (Evans, 1986). The resulting imbalance in the focus of inquiry has left politics lopsided in its approach to gender. Women, studied by a small but growing number of academics, attract far more attention than 'men', who are studied by almost no one as a gender. The lack of investigation into the effect of being a man on political action has left investigations of the 'gender gap' lacking half of its analytical base. This paper will strive to begin the process of solving that oversight in Canada, by constructing a framework through which the block tendencies of both male and female voters can be measured. Carver wrote that " i f we fail to take it upon ourselves to scrutinize men and masculinity in depth, in tandem with our more sophisticated and nuanced understandings of women and 'the feminine,' then we shall severely restrict our own critical analysis and understanding of the gender order, as 'gender is not a synonym for women'" (Carver, 1995. 28-29). This is especially true of research on gender and elections, where the focus on a women's voting block has -1-ignored the role men have had in electing conservative governments. Considering men as a viable voting block will allow for a more realistic appraisal of the relative power of each gender in questions of government formation. Studying men as men has not been a major part of political science research, but inquiries into the causes of male voting behaviour may help to explain the rise of radical right parties and the current transformation of the left. This thesis will be structured in five sections. The second chapter of this thesis will be a layout of the history of gender in political science, criticising the approach of traditional political scientists before the feminist revision. The modern gender gap in Canada across the past twenty-five years will be presented with a focus on the changing character of that gap. Chapter three will use multiple regression analysis to explore beneath the surface of the contemporary gender gap in Canada and determine what issues push men and women apart at the polls. Chapter four will critique the existing bias in the literature on the gender gap, linking oversights in feminist theory to current research. Chapter five will use rational choice theory to construct a model of power in voting blocks and prove that men are increasingly important in elections. The sixth section will explore potential causes of male voting behaviour, conclude the discussion of the gender gap and produce recommendations for future research. Before continuing, perhaps it would be best to declare what I am not attempting with this paper. First, in no way am I refuting the contention developed in feminist theory that gender is a significant structural feature in the construction of hierarchies of power and a key variable in understanding social relations. Second, while I am a male researcher investigating the effect of masculinity, I do not want to be confused with writers who may advocate that only women can -2-write about women and men writing about men. Third, the point of my work is not to defend men's rights or modernise hegemonic masculinity by helping men to cope with feminism without actually relinquishing power. The oppression of women by men is taken as fact in this work, although the author rejects the notion that all men are always oppressive, collectively and individually. Gender will be defined in this study as the essential and learned behaviour that accompanies physical sex differences. This definition will classify gender as a dichotomy and ignore many of the complexities involved. Sociologists, biologists and gender theorists have often understood gender among humans to be "a continuum of norms and behaviours socially constructed, socially perpetuated and socially alterable" (Gould and Kern-Daniels, 1977. My italics). Even some feminist and political science scholars are critical of the oversimplification of gender and celebrate the diversity of sexuality, experience and context that modify and crosscut genders (hooks, 1981; Orr, 1997). But for the purposes of this investigation, the assumption will be made that gender is a mutually exclusive cleavage that places every community member into one of two groups1. While this limits the vision of the investigation, hopefully it will be made up for in brevity and a better analysis. i The ambiguities that surround gender involving chromosomal variation, morphological difference, and the complexities raised by homosexuality, bisexuality, transgenderedness, cross-dressing, etc. will be suppressed due to a simple lack of space. It should be noted that this cleavage does not imply gender unity in belief and action, nor complete distinction in any factor other than sex. -3-Chapter 2 - Just What is this Gender Gap Thing?: "As the faculty of writing has chiefly been a masculine endowment, the reproach of making the world miserable has always been thrown upon the women." Samuel Johnson Being human unavoidably immerses one in gender politics, in a struggle of power, tradition, institution, biology, chemistry and emotion. Almost daily, another salvo is fired in the great debate surrounding the nature, cause, and effect of gender-based difference. Men may be from Mars and women from Venus, but here on Earth gender is one of our most salient cleavages (Johnston et al, 1995). Gender has been identified as an influence on a variety of political attitudes and behaviours. Some of these observations appear durable and robust. For instance, male respondents usually appear to be more supportive of the use of force by the state than women in both domestic and international situations (Norris, 1988; Warner, 1998). Other arguments appear to be more fleeting, such as a significant difference in participation rates between the genders in elections (Campbell et al, 1964). To comment on the current analysis of the gender gap, the roots of that analysis should be explained. The first section of this chapter will outline the traditional approach of pre-feminist political science toward questions of gender and the challenges to this orthodoxy from feminist scholarship. The second section will investigate the movement of the gender gap in Canada across time, searching for patterns and shifts between the two genders and various parties. Some hypotheses will be suggested to explain the relationship between gender based voting difference and the politics of government formation. The current state of the gender gap will be debated and potential future trends considered. -4-A Short History of Gender in Political Science: Gender currently stands alongside region, religion, language, education, ethnicity and class as an identifiable demographic characteristic linked to political altitudes. Many feminists argue that "the dichotomy of masculine-feminine is a structural feature of social life, rendering gender an analytical category with systemic implications for advancing our understanding of social relations" (Peterson, 1998:18). While the dichotomous nature of gender may be open to challenges, gender's capacity as an analytical tool has gone from almost universally ignored to almost universally accepted. This shift required a difficult reconceptualization of many of the assumptions of the discipline. But change has produced better research, cognisant of gender and its effects. Gender is now seen to be a major factor that must be controlled for and addressed in research. Despite the current acceptance of the importance of analysing politics and gender, the early legacy of political science with respect to gender is primarily constructed of negligence, ignorance, and unreflective stereotyping. The effect of gender on the political attitudes, orientation and behaviour of individuals and groups was rarely considered as a variable in political inquiry. Women, when acknowledged at all, were usually considered 'uninteresting' by political scientists earlier in this century (Rinehart, 1992). The rare analyst that did consider 'woman' was often willing to accept stereotyped characteristics without rigorous empirical testing, such as identifying the conservative power of maternity as the reason for women's conservative voting patterns. Men were not even thought of as a gender, since 'man' was the principal subject for academic inquiry. But, as later chapters will demonstrate, the universal 'man' and the gendered 'men' are different. To emphasize further, many classic works of political science abandoned scientific methods when discussing gender and allowed normative beliefs to guide their conclusions. Almond and Verba's Civic Culture contains this unsupported passage: "Whenever the consequences of women's suffrage have been studied, it would appear that women differ from men in their political behaviour only in being somewhat more frequently apathetic, parochial, conservative and sensitive to the personality, emotional, and aesthetic aspects of political life in electoral campaigns"(Almond and Verba, 1963:140). In 1970, Van Loon commented on women's lower participation in the 1965 general election in Canada, writing that "in many families, politics is still considered to be a male activity with the husband attempting to guide the wife's vote but hardly vice-versa"(Van Loon, 1970:389). Pomper wrote in Voter's Choice that "there are few issues that are primarily related to a person's gender... Nor is social segregation between the sexes likely as long as hormones do their pleasant work." (1975:84). Duverger's work stands as a notable exception to an otherwise disappointing history (1955). But even he failed to question the prevailing definitions that excluded women from the corridors of power (Lovenduski, 1981). Reviewing electoral studies of gender before 1980, we find evidence of gender-based voting difference in many countries, although the reasons may be more complicated than the contemporary scholars had thought (Inglehart and Norris, 1998). Men were more likely to be found voting for left-wing parties like the British Labour Party or the Italian Communists than women, who displayed a tendency to support centre-right parties like the British Conservatives or Italian Christian Democrats (Duverger, 1955; Campbell et al., 1960). The implication that this 'traditional voting gap'1 was caused by women being more conservative in their political attitudes and values is more difficult to prove, but was commonly held. In traditional political science research, women were perceived to be aping their husbands or, where they did differ from men, adopting more conservative stances because they did not vote for leftist parties (Courtney and Smith, 1966; Laponce, 1969). Most of the early explanations relied on presumed differences in religiosity, longevity and labour force participation to explain women's conservatism: "It is often assumed that women's stronger ties to the family, neighbourhood and church make them more conservative, while the predominance of men in trade unions leads them to support the left" (Norris, 1988:219). Adrienne Rich wrote that "certainly [Patriarchy] has created images of the archetypal Mother which reinforce the conservatism of motherhood and convert it to an energy for the renewal of male power"(Humm, 1992:273). Women were perceived to be the preservers of home and hearth, defenders of children and morality, and the keepers of faith. Describing voting women as either passive, emotional imitators or cautious conservatives reinforced established assumptions about gender and allowed researchers to move onto what they saw as more 'interesting' things. The methodology and conclusions of these early works have been challenged extensively by more recent scholarship. While traditionally the lack of support for the New Democratic Party among women was used as evidence for women's passive conservatism, more recently these findings have been attributed to the failure of the labour movement to mobilise i The terms 'traditional voting gap' for a gap with women to the right and men to the left and 'modern gender gap' for a gap with women to the left and men to the right come from Inglehart and Norris, 1998. -7-in traditionally female industries and integrate women into the class networks through which political messages are conveyed (Kopinak, 1987). Bourque and Grossholtz convincingly argued that the American literature on the subject assumed women to be essentially passive and feminine, that male political dominance was a fundamental good in society, and demonstrated that the 'evidence' provided to support these beliefs was shoddy, i f not outright contrived (1974). McCormack similarly concluded that such biases had been methodologically extended in the failure to apply appropriate controls (1975). By employing such controls, a later study found the much vaunted 'participation gap' to be the illusionary product of differences in socio-economic factors, such as education, age and income, rather than evidence of women's natural passivity (Poole and Zeigler, 1985). As more gender-conscious researchers entered the discipline, many of the gross oversights and rampant chauvinism of the earlier scholarship were gradually amended. Stimpson dubbed this 'the first step in the new scholarship on women' (1984). While this change was obviously essential, Stimpson's laurel should acquaint the observer with the raison d'etre of many of these reformers: adding women to research in political science. The movement to introduce women into political science had difficulty overcoming ingrained attitudes until a dramatic shift in the political electorate made gender an undeniable factor in voting (Lovenduski, 1981). The event that brought the political behaviour of gender to the forefront of both scholarly and popular thought was the 1980 American presidential election in which Republican nominee Ronald Reagan defeated the incumbent Democrat, Jimmy Carter. Gender was undeniably a significant factor in this election for several reasons. First, women participated in greater numbers than men, permanently countering the argument that women are naturally uninterested in politics (Frankovic, 1982). Second, issues of gender and the role of women were at the centre of the debate, especially the Equal Rights Amendment (ERA) that Carter supported and Reagan opposed (Abzug, 1984). Finally, 55 per cent of male voters supported Reagan, compared to 36 per cent who supported Carter. But among women, support for Reagan and Carter was split 46 to 45 (Frankovic, 1982). The difference in voting behaviour in 1980 was larger than any seen before between men and women, somewhere between 9 and 12 percent, depending on the measure. Also, it was the reverse of the traditional gender gap, with men supporting the more conservative candidate in greater numbers. In short, 1980 was the first documented appearance of the 'modern gender gap'. Table 2.1 documents that election's results. Table 2.1: 1980 U.S. Presidential Election Results divided by Gender: Candidate Total Vote Female Vote Male Vote Gap (? - <f) Ronald Reagan (Rep.) 50.5 46 55 -9 Jimmy Carter (Dem.) 40.5 45 36 9 John Anderson (Ind.) 9 9 9 0 Data from CBS/New York Times survey. n=l 5,201. The Canadian Gender Gap: While the election of Reagan dramatically demonstrated the importance of gender in politics, a 'modern' gender gap quietly helped to decide Canadian elections earlier than in the U.S. The right-of-centre Progressive Conservatives drew more men than women supporters in elections held between 1974 and 1980. The support of the centrist social reform Liberal Party drew predominantly from women voters, who were more likely to support Pierre Trudeau than contemporary men. The left-wing New Democratic Party (NDP) -9-drew poorly among women in the 1970's, a phenomenon that disappeared with the beginning of the 1980's. These shifting patterns are noted below in the election results for the period of 1974-1988 taken from the Canadian National Election Studies (CNES). The gap is calculated as is most common: party vote from women minus party vote from men. Positive results indicate more women supporting a party and negative results indicate more men. This data has been refined by a method reported in Appendix One in order to more accurately reflect the actual level of voting during this period. Table 2.2 Canadian Federal Election Results 1974-1988 divided by Gender (Refined Data): Year NDP Men NDP Women NDP Gap Lib. Men Lib. Women Lib. Gap PC Men PC Women PC Gap 1974 16.9 13.1 -3.8 38.6 47.4 8.8 37.5 32.5 -4.9 1979 18.1 17.8 -0.3 36.1 43.9 7.8 40.6 31.4 -9.3 1980 19.2 20.8 1.6 39.4 48.5 9.1 38.9 27.1 -12.2 1984 19.0 19.0 0.0 26.3 29.7 3.4 51.5 48.4 -3.1 1988 19.5 20.5 1.0 28.7 35.3 6.6 46.8 39.2 -7.6 Data from Wearing and Wearing, 1991. Refined Data (See Appendix One) This early appearance of the modern gender gap may be due to the policies of the government. The welfare liberalism1 of Trudeau enjoyed enormous support among women, great enough to offset the Conservative's advantage among men. In a similar vein, Trudeau's Just Society programs and flamboyant personality drew women voters who might otherwise 2 Welfare Liberalism: A concern for individual freedom that believes state intervention in the economy can give that freedom more effective worth (Campbell, 1996). -10-have supported NDP policies. Thus, the early development of a modern gender gap in Canadian politics should be attributed to Trudeau, a legacy symbolic of his distinct approach to government. Had a business liberal3 like Robert Winters or John Turner been selected at the 1968 convention, it is likely the contemporary shift of women to the NDP would have happened earlier and the Liberals would have had a more difficult time winning in the 1970's. With Trudeau's departure and the selection of former finance minister John Turner as leader, the Liberal base among women began to dissolve. Turner's bum-patting of Liberal Party president Iona Campagnollo and his conservative rhetoric about spending and the deficit opened the door for many women to move to the Tories as part of a general shift. The 1984 election saw women almost as likely to support Brian Mulroney's Progressive Conservative party as men. This supports the idea that Trudeau was essential for the modern gender gap the Liberals drew strength from. However, in the free trade election four years later, the modern gender gap returned to Canadian politics with both the Conservatives and Liberals drawing significantly more support from their traditional gender. Turner's fierce opposition to the Free Trade Agreement and defence of Canada's welfare state helped to draw some women back to the Liberal party (Gidengil, 1995). The Canadian party system fragmented between the 1988 and 1993 elections. The Conservative coalition forged by Brain Mulroney in 1984 dissolved into three factions: the sovereignist Bloc Quebecois, the neo-conservative Reform Party, and a Tory rump remaining in the Progressive Conservative Party. This shift in the party system will be an inconvenience, 3 Business Liberalism: A concern for individual freedom, focused on property rights, that views the state as the principle threat to that freedom (Campbell, 1996). -11-but its impact on analysis of the gender gap is relatively minor. The Bloc Quebecois was the only new party to correspond to the classic model of realignment by mobilization, while the Reform party grew mainly by converting former Tories to a new vehicle (Johnston et al, 1996). The Reform surge may be interpreted as a fractionalization of the 'conservative ideological family', with men still dominating that family's voters (Bartolini and Mair, 1990). The Bloc Quebecois is a gender neutral party that does not enter into discussions of the gender gap in Canada. So while the conservative vote may be split, the basic ideological family remains male-dominated. Table 2.3 Canadian Federal Election Results 1993 and 1997 divided by Gender: Gender NDP Liberal PC Reform BQ Male 1993 6.2 40.3 13.2 22.3 14.5 Female 1993 7.6 44.1 15.2 16.3 13.1 Gap 1993 1.4 3 g** 2.0** -6.0*** -1.4 Male 1997 8.1 37.1 16.3 23 9.3 Female 1997 13.1 37.4 15.7 15.7 8.8 Gap 1997 5.5** 0.3 -0.6 _7 3*** -0.5 **p < 0.01, ***p < 0.001, 1993 n=4672. 1997 n=1840. 1993 data refined. The gender gap was not as large in 1993 as it has been on occasion in the past in Canada. No party was so divided in the gendered base of their support as the 12 point difference experienced by the Conservatives under Clark in 1980 or even the moderate differences experienced by both the Liberals and the Tories in 1988. Reform was the only party to produce a gender gap that was significant to (p > .001). Both the Liberals and the Tories drew more women than men, and were significant to (p > .01). The gender gap for the Tories -12-was the reverse of earlier years, with Kim Campbell attracting (or retaining) more women than men. The 'conservative family of parties' retained its male-dominated voter base, with a combined vote of 35.5 per cent among men and only 31.5 per cent among women. While the NDP and the Bloc Quebecois garnered slightly more of their support from one gender than the other, the gap they experienced was not significant.4 Between 1993 and 1997, the Liberals lost some men and significantly more women from their coalition. These losses among women may be attributed to the 1995 Budget and the Liberals conversion to deficit fighting, as Liberal support among women voters seems to be somewhat dependant on their economic positions. When the Liberals take welfare liberal positions, they do well among women (1974 - 47.4 %, 1980 - 48.5%, 1988 - 35.3%, 1993 -44.1 %. Average: 43.8%). When anti-deficit, anti-spending positions are taken, the Liberal vote among women declines (1979 - 43.9%, 1984 - 29.7%, 1997 - 37.4%. Average: 37%).5 Figure 2.1 illustrates the shifts in support for the Liberal Party over the period in question. Women move away from the Liberals more than men in 1979, 1984 and 1997, each time diminishing the gender gap. When Liberals change from being financial doves to financial hawks, they lose more women supporters than men. 4 While support for sovereignty is significantly divided by gender, with men more supportive of independence, support for the Bloc Quebecois is gender neutral (Johnston et al, 1996a). The ability of Lucien Bouchard to appeal to moderate francophone women, the social democratic base of the party, and the distinction between voting in elections and referenda may account for this discrepancy. 5 These three elections followed either spending reductions to lower the deficit (1979, 1997) or statements by the party leader that such actions would have to be taken (1984). (Chretien, 1986; Barlow and Campbell, 1995; Valpy, 1998). -13-Figure 2.1: The Gender Gap in Support for the Liberal Party 1974-1997 Liberal Vote Among Women Liberal Vote Among Men The present government's break with Trudeau's Social Justice legacy and the embracing of neo-liberalism have cost the Liberal Party Trudeau's gift of disproportionately high support among female voters. If the Liberals retain their new right-of-centre image, the result could be a net deficit in support among women. The corresponding growth of the NDP, especially among women, seems to confirm this. The NDP's support among women grew dramatically in both 1979 and 1997, and remained stable in 1984 despite Mulroney's landslide victory. However, a business liberal position may help staunch the flow of males to the conservative family of parties. While the Liberals did lose support among men between 1993 and 1997, they may have lost more women supporters and, especially, more men had they not moved to address the deficit. -14-In 1997 the Reform party became an even more significantly male than female movement, with increasing support among men and losses among women. The replacement of Tory leader Kim Campbell with Jean Charest attracted more men to the party, erasing the earlier female-heavy gap. The 'conservative family of parties' was more male dominated in 1997. While the total number of women voters voting for conservative parties declined by 0.1 per cent, the total number of men increased by 3.9 percent to 39.3 per cent of the male vote. Combined, these conservative parties possess a 7.9 per cent male-heavy gender gap and attracted more male voters than the Liberals. The Bloc Quebecois remained a gender neutral party. The New Democratic party coalition became more female-dominated, drawing significantly more women than men for the first time. Figure 2.2: The Gender Gap in Each Canadian Party 1974-1997 0) CO E CD UL V Q) CO CO E a> LL A 0 CO BQ Gap Liberal Gap NDP Gap PC Gap Reform Gap 1974 1979 1980 1984 1988 1993 1997 -15-The Figure 2.1 combines the earlier tables in a graph form to present the changing gender gap in support for each party over time, demonstrating trends over the past quarter century. It illustrates the key role gender has played in the formation of governments in the past. However, gender difference may be less important to the question of federal government formation at present. The gender gap in the 1970's was between the Liberals and Conservatives, the two parties that traditionally formed governments. The gender divisions appear to be evening out in these two moderate parties and widening for the more extreme parties on the left and right. Gender has become less important to the central question of government formation, although it remains an important factor in determining whether the government will be a majority or a minority. The second trend noticeable in the table is that the conservative family of parties show a larger gender gap at present than do the centrist Liberals, the leftist NDP or even the two centre/left parties combined. As the Liberals move to a more pro-business position and reject welfare liberalism, the result is a 'gender neutral' governing coalition. If Preston Manning is successful in his attempt to reunite the conservative family of parties, a male-dominated coalition may again contest to form a government. Already we see male-dominated conservative coalitions in power in Ontario and Alberta. The possibility of a male-based conservative party competing with gender-neutral Liberals federally begs a number of questions about how gender influences politics and who really benefits from the gender gap. A final observation on the history of the gender gap provokes the question: if the gender gap benefits women, and women are more left-wing, why have parties and policies moved to the right since the early days of the modern gap? The Liberal party is barely recognisable today -16-when compared to the social justice and Keynesian economics of the Trudeau era (Liberal Party, 1997;.Bashevkin, 1998). The Progressive Conservative party has shed its Red Tory wing and embraced a credo of tax cuts and budget slashing (Progressive Conservative Party, 1997). The Reform party is far more successful than its Social Credit predecessor, despite a similar platform of social conservatism and Western discontent (Reform Party, 1997). Even, the leader of the nominally socialist NDP has called for the party to reach out to business interests and embrace the 'Third Wave' centrism of Tony Blair. A investigation of the underlying structure of the gender gap must be made to determine the significance, direction and nuance of gender-specific attitudinal differences. If men are more right-wing than women, both in attitudes and in voting, then men must be a factor in drawing the electorate increasingly to the right. -17-Chapter 3 -What Lies Under the Surface?: "If you can fill the unforgiving minute with sixty seconds' worth of distance run, Yours is the Earth and everything that's in it, And - which is more - you'll be a Man my son!" Rudyard Kipling The history of the gender gap in Canada provokes as many questions as it answers. Are men actually more conservative than women, or do we need to look for other possible causes of voting difference, such as comparative mobilization strategies or the differing appeal of party leaders? If men are actually more conservative than women, are they more culturally conservative, more fiscally conservative, more authoritarian, or all three? Why are men more conservative than women? Which gender decides who forms the government? Is there a male backlash causing the rightward movement of Canadian parties? This chapter will employ quantitative survey analysis and an exploration of the existing analytical frameworks to answer some of these questions. Scholarly literature on the gender gap will be summarized in this chapter to begin the process of determining what lies at the root of gender based voting difference. The literature review section will point to political ideology as central to the gender gap, with socialisation, structural and situational constraints as contributing factors to attitudinal difference. Ideological dissimilarities between the sexes are widely held to be key intervening variables between gender and party vote. The second section of this chapter will investigate the differing political ideologies of men and women. If men are actually more conservative than women, this conservatism will be observed in underlying value structures. With multivariate regression analysis, we can test the effect that a series of ideological scales has on the gender -18-gap in Reform party support. This will allow a determination of the type of differences between men and women that create the gender gap. Identifying the type of attitudinal difference between men and women will help develop theories to explain voting difference. Reform was selected for this study because it is the party with the largest gender gap in Canada. Differences will have greater significance because of the wide division between men and women over Reform. However, a similar investigation could have been made of the gap in New Democratic Party support. It is important to stress that the findings here are only valid for supporters of the Reform party in 1997. The underpinnings of the gender gap found in other parties in 1997 or in earlier years may be very different. The Literature in Review: During the era of the traditional voting gap, and in the present era of the modern gender gap, a variety of theories were developed to explain gender voting difference. There are four established approaches to analysis of gender difference: • Political ideology and attitudes, • Structural or demographic effects, • Situational or lifestyle effects, And early childhood socialization and essential difference. The political ideology approach is the most common of the four and attempts to link the increasing divergence in male-female attitudes toward key political issues with the changing party loyalties of each sex. Attitudes towards the use of force and support for compassionate public policy are held to account for the predominance of women in left/centre parties by a legion of authors (Frankovic, 1982; Goertzel, 1983; Klien, 1984; Conover, 1988; Cook and -19-Wilcox, 1991; Wearing and Wearing, 1991; Rinehart; 1993). Gender causes these attitudinal differences to occur, and it is these ideological dissimilarities that in turn cause voting difference. But political ideology is itself the result of other factors in people's lives. The other three approaches attempt to explain gendered division of opinion, and look at what creates differing attitudinal perspectives in men and women. There is some interdependency between these factors, as, for example, socialization of women toward child-rearing and nurturance leads in turn to greater situational effects from raising children, which may in turn lead to the structural effect of lower wages due to seniority lost when caring for children. Gender is a powerful factor in our lives, and the threads of its effect are tightly interwoven. Separating them is a difficult task. Structural factors are demographic characteristics that are affected by a subj ect's gender. To provide some examples, structural factors include the differences in average levels of education, social class, religiosity, workforce participation, and welfare state dependency of each gender. For instance, the increasing 'feminisation of poverty' means that women are more often in need of the programs of the welfare state to feed and shelter themselves, to be in a position to doubt the worth of capitalism, and to support affirmative action or pay equity legislation (Gidengil, 1998). Women are more likely to rely on the state for aid, to be the ones delivering that aid as social workers and bureaucrats, and more likely to oppose cuts to that aid (Erie and Rein, 1988). Differences between women and men in this school of thought are considered to be strictly circumstantial and contingent on the economic situation in which each was likely to be found. Proponents of this position often subscribe to the liberal feminist -20-argument that once women have gained full-employment and integration into the market society, gender differences will decline (Togeby, 1994). Situational constraints include women's institutional role as care-giver to children, the elderly, and disabled people. These are thought to instill a greater appreciation for social-welfare policies, personality traits of caring and nurturance, and an understanding of the precariousness of life (Anderson and Cook, 1985). Situational effects help to explain why wealthy women are as likely to support welfare state programs as poor women. Women of all classes participate in child-rearing and care-giving to a greater extent than men (Jump, 1987; Tavris, 1992). Societal expectations are such that women are much more likely to feel obligated to come to the aid of those who require nursing and care than men in a similar position. Political parties that threaten programs to help care for children or the elderly will find themselves not just losing votes from day-care workers or seniors, but from women as well. 1 The ethics and beliefs forged in nurturing roles can colour a variety of political attitudes, from war to poverty. Socialization-based analyses of the gender gap focus on the early life experiences of men and women, especially the adoption of 'gender roles' that differentiate the sexes on a level the subject may not even be aware of (Gilligan, 1983; Sapiro, 1983). Alternatively, these differences may be essential, and the result of biological variations between men and women (Chodorow, 1978; Okin, 1990). O'Neill has demonstrated that biological difference or early life socialization account for at least part of the gender gap (1996). She attributes these differences i Observers from Amy Schwartz of the Washington Post to Phyllis Schafly of the Eagle Forum agree that it was the Republican's brutish stand on Medicare and Medicaid programs for the elderly that most turned off women voters in the 1996 American election. (Schwartz, 1996. Schafly, 1996). -21-to distinct political cultures for men and women2. Socialization analysis stresses the existence of fundamental differences between men and women, both in opinions and how these opinions are clustered and structured. These differences are thought to be distinct from other structural or situational differences, and would persist after economic or legal equality for women was achieved. Judging from the literature, the logical progression to clarify voting dissimilarity is to first identify the attitudinal and ideological political differences undergirding the gender gap. After the issues that contribute significantly to the gender gap have been identified, an examination of the underlying situational, structural and socialised differences may be undertaken. This method will allow a detailed accounting of those issues that produce gender variation and will determine the answers to several of our questions in two ways. First, it may be determined i f men are in fact more conservative that women, or if these effects are brought on by an intervening variable. Second, i f men are found to be more conservative, the ways in which their conservatism manifests itself are of importance. By determining what issues men and women differ over, our study will be in a better position to investigate the political behaviour of men and determine the causes of the current gender gap structure. Political Ideology and the Gender Gap: Our goal in this study is to determine what ideological indexes account for significant portions of the gender gap. First, we will regress the 2 Inglehart defines culture as a "system of attitudes, values and knowledge that is widely shared within a society and transmitted from generation to generation" (1990:18). O'Neill posits that long-standing differences in opinion between women and men is the result of distinct political cultures, with women's political culture based on pragmatism and problem-solving rather than power and conflict. This culture is held to be in transformation as younger women discard religious ties and embrace more liberal positions than their elders (1996). -22-dependent variable of Reform Party Support from the 1997 election on an independent variable of gender. This will determine the size and composition of the gender gap in the Reform Party. Subsequent three-variable regressions will be performed to determine independently the effect of a single intervening ideological variable on the relationship between gender and Reform party voting. Of minor interest to us is the R 2 , with a high level indicating a strong relationship between that attitude and Reform Party support. But our interest is not Reform party voting per se, but the difference between the genders in voting for Reform. An index can have a large impact on Reform, measured by a large R 2 , but almost no impact on the gender gap in Reform party voting. What will measure the significance of an ideological index on the gender gap is the shift in the coefficient on 'gender'. If the size of the Gender coefficient decreases when an ideological scale is added to the regression, then the second covariant is drawing explanatory power away from gender to itself, indicating that the difference of opinion on that issue cluster is partially responsible for the gender gap. Figure 3.1 illustrates the relationship being modelled. Figure 3.1 Graphic Representation of Model Testing Attitudinal Intervening Variables and the Gender Gap in Reform Party Support Gender Various Intervening Attitudes Gender Coefficient Reform Party Support -23-Each regression consists of three variables: Reform Party Voting, Gender and one of the attitudinal indices being tested. As each regression is performed, the data reported will be the R 2 , to measure the index's impact on Reform, the Gender variable coefficient, to indicate Gender's impact on Reform, and the statistical significance of Gender. If the Gender coefficient falls dramatically, that index has drawn a significant portion of the explanatory power of Gender. Therefore, that variable is an important cause of the gender gap in Reform Party support. If the Gender coefficient in not diminished by the introduction of the control for this attitude than it has little significance in the gender gap. The standard error for the Gender variable is always .018, and so is not reported independently for each variable. The ideological scales will be grouped for display purposes into economic, cultural and authority issues, but each scale will be regressed independently. The economic indices are Capitalism, Continentalism, Anti-Welfare and Macroeconomic Policy. The cultural indices are Moral Traditionalism, Feminism, Anti-Ethnic and Anti-French. The economic and cultural indices employed were all modelled on existing indices developed by the 1993 CNES team3. The authority indices for Populism, Alienation, and Use of Force were developed independently using questions from the 1997 Canadian National Election Study. These indices were added to the established set to test potential or established theories that they may help to cause the gender gap (Norris, 1988; Warner, 1998). Each index uses three to eight questions from the 1997 CNES and measures the combined level of support for all of the questions'in each index. See "The 1993 Canadian Election: Realignment, Dealignment or Something Else?" (Johnston et al, 1996). -24-The individual questions and SPSS syntax used to create these indices can be seen in Appendix One. Table 3.1:The Effect of Significant Issue Indices Regressed on the Gender Gap in Reform Party Support (Three Variables): Variable R 2 Coefficient of Gender Significance of Gender Gender (Two-Variable) 0.008 -0.0762 .000*** Effect of Economic Issue Scales on Gender Gap Capitalism 0.050 -0.0436 .017* Continentalism 0.019 -0.0553 .003** Macro-economics 0.019 -0.0592 .001*** Anti-Welfare 0.012 -0.0656 .000*** Effect of Cultural Issue Scales on Gender Gap Moral Traditionalism 0.087 -0.0561 .001*** Feminism 0.061 -0.0400 .027* Anti-French Opinion 0.071 -0.0607 .001*** Anti-Ethnic Opinion 0.053 -0.0635 .000*** Effect of Authority Issue Scales on Gender Gap Populism 0.016 -0.0795 .000*** Alienation 0.040 -0.0815 .000*** Use of Force 0.061 -0.0264 .151 Dependent Variable = Reform Party Support. n=1840. *** p<.001, **p<.01, *p<,05. Economic Indices: Of the economic scales, only Capitalism and Continentalism registered a significant impact on the gender gap, by lowering the significance of gender below -25-99.9% and diminishing the gender coefficient substantially. The Continentalism scale is composed of questions of liberalised international trade, especially trade with the United States. Continentalism was expected to have an impact on gender, given Gidengil's contention that "embodying as it did an individualistic, competitive, market-oriented world view, the free trade issue pushed to the fore questions about the appropriate role of markets and the nature of the welfare state" (1995: 387). However, the effect of Continentalism is weak when compared to Capitalism, and its effect on gender in a multiple regression with Capitalism demonstrated that its impact was collinear with that of Capitalism. In Table 3.2, we see that Continentalism's explanatory power is greatly diminished when Capitalism is regressed at the same time. Continentalism's impact is surprisingly low and it will be set aside as an insignificant variable. Table 3.2 :The Effect of Significant Issue Indices Regressed on the Gender Gap in Reform Party Support (Four Variables): Variable Significance Coefficient Standard Error Gender .040* -0.0378 0.018 Capitalism .000*** 0.387 0.048 Continentalism .047* 0.0547 0.047 Dependent Variable = Reform Party Support. R2=.051. n=1840. *** p<.001, **p<.01, *p<.05. While support for the welfare state has long been held as a major issue for women in North America (Erie and Rein, 1986; Carroll, 1988), attitudes toward welfare programs do not correlate significantly with the gender gap in Reform party support. The gender coefficient was almost completely unchanged. Macroeconomic policy did not significantly affect the gender gap in 1997. Both women and men supporting the Reform party seem equally committed to a deficit-fighting agenda. However, it should be pointed out that more than 70% of Canadians -26-in this survey adopted the most anti-deficit and anti-debt position possible and so the index does not explore the nuance of opinion. Capitalism, a scale measuring support for the ideals of the free market rather than socialism, had the greatest impact of any economic variable on Reform Party voting. The Capitalism index also had the greatest effect on gender, lowering the significance below 99% and lowering the gender coefficient by more than a third. Capitalism was the only economic variable to explain a significant part of the gender gap. The fact that significant difference is contained only in attitudes toward Capitalism, and is not seen in other economic indices, indicates that differences between men and women are not based on more specific areas like welfare, trade and spending, but in general attitudes toward the market. More women than men believe in an activist state and that government has an obligation to ensure fairness and equal opportunities. Cultural Indices: Al l of the cultural indices in this series of regressions had a major impact on predicting Reform party support, far more than any of the economic issues. However, only Feminism had a major impact on the gender gap, basically halving the gender coefficent. The Feminism index was built from questions exploring opinions on job quotas for women, current hiring practices, marital violence, and the goals of the women's movement. It tests attitudes toward the feminist movement and the present status of women in gaining equality with men. Feminism had an even greater impact on the gender gap than capitalism, demonstrating a definite anti-feminist bias amongst male Reform party supporters. -27-Moral Traditionalism, an index measuring commitment to conservative morality on issues like homosexuality and traditional patriarchal family structure, was the single greatest predictor of Reform party support, with an R 2 of 0.087. However, Moral Traditionalism was a poor covariant of gender, with only minor reductions in the gender gap. Earlier research has demonstrated that women are, as a group, more conservative than men on issues of morality (O'Neill, 1996). Women's greater commitment to religion fosters conservative attitudes on questions on personal morality. So it is not surprising to see that moral traditionalism had little impact on repulsing women or attracting men from Reform. However, Moral Traditionalism's power in explaining support for the Reform party in general reinforces beliefs that it is a party of social conservatives. Negativity toward new Canadians, aboriginal Canadians and minorities was compiled into an Anti-ethnic index that registered a significant impact on the Reform vote, but a negligible impact on the gender gap. Anti-French sentiment, a compilation of complaints against bilingualism, the Quebec government, and French-Canadian nationalism, registered a major impact on the Reform vote, but again did not affect the gender gap. While nativist sentiment correlates highly with Reform Party voting, women and men appear to be equally attracted or repulsed by the Reform Party's perceived stand on issues of ethnicity or language. Of all cultural scales, only differences of opinion over feminism help to explain the gap. Authority Indices: The three authoritarian indices all produced surprises. Populism is central to the rhetoric of the Reform Party, and both Reform and the NDP claim to be populist political parties. Perhaps the gender gap has translated itself into variants of populism, with -28-populist males supporting Reform and populist women voting NDP. Unfortunately for this theory, Populism, and its alter-ego, Alienation, were complete failures in explaining the gender gap. The Populism index measures support for grassroots decision-making and referendums, but failed to make a major impact on either the Reform vote or gender. Alienation, a measure of satisfaction in Canadian representative democracy and the federal government, explained a part of Reform voting, but was even less of an impact on the gender gap. Populism recorded one of the lower R2s of the exercise. Alienation from the federal government was a better indicator of Reform support. Reform supporters appear to be far less 'anti-elite' than 'anti-federal government.' However, while Populism and Alienation were poor co variants, Use of Force issues were the single largest covariant of the gender gap explored. Its effect was almost twice that of either Capitalism or Feminism. It was the only variable to render the gender coefficent insignificant and lowered that coefficent by nearly two thirds. This scale measures attitudes toward violence domestically (capital punishment, gun control) and internationally (military spending, foreign intervention). Use of Force was also a key predictor of Reform Party Support, tied with Feminism as the third strongest R2. Obviously, the stance the Reform Party has taken on these issues has alienated women voters, while attracting a disproportionate number of men. This is consistent with other studies which have found Use of Force issues to be the largest differences between male and female opinion (Warner, 1998) and a key cause of the gender gap in support for Ronald Reagan in 1980 (Mansfield, 1983). -29-Relative Explanatory Power: The ideological issue scales test produced three variables with significant explanatory power when regressed on the gender gap: Capitalism, Feminism and Use of Force. The next regression series will measure the combined effect of these variables on the gender gap, to determine which variable explains the largest portion of the gender gap and if their effects are collinear or distinct. A five-variable regression will be employed to test the combined effects of all three significant variables on the relationship between Gender and Reform Party Support. The R 2 is not as important in this regression, as we have already determined the relative strength of the variables in predicting Reform Party support, and as this is only of secondary interest to us. Rather the findings will focus on the power of each variable to influence the significance of gender. Table 3.3 The Effect of Significant Issue Indices Regressed on the Gender Gap in Reform Party Support (Five Variables): Variable Significance Coefficient Standard Error Gender .696 0.00 .018 Capitalism .000*** 0.284 .040 Feminism .000*** -0.314 .046 Use of Force .000*** 0.312 .039 R 2 = .117, ANOVA sig = .000, Standard Error of the Estimate = .3708, n = 1840. *** p<001. Dependent Variable = Reform Party Support. As can be seen in Table 3.3, these three indices explain the gender gap, reducing it to utter insignificance. If women and men shared the same opinions on issues of Capitalism, Feminism and Use of Force, there would be virtually no gender gap in support for the Reform -30-Party. The R 2 is strong for these three variables at .117, indicating that all three variables are important influences of Reform Party support in their own right. However, this table gives us no understanding of the relative impact of each variable on the gender gap. In order to test the impact of each variable relative to the other two significant indices, a series of four part regressions will be run. Each regression will involve Reform Party support as the dependent variable and Gender as an independent variable. Each will also include two of the significant variables so the excluded variable may be understood. If the significance of Gender rises, the excluded intervening variable is a strong cause of the gender gap. If its remains low, then the excluded variable is less important, but still significant. Table 3.4: The Effect of Significant Issue Indices Regressed on the Gender Gap in Reform Party Support (Four Variables): Variables Included Variable Excluded R 2 Gender Coefficient Gender Significance Feminism/Use of Force Capitalism 0.099 -0.005 0.783 Capitalism/Use of Force Feminism 0.087 -0.009 0.605 Feminism/Capitalism Use of Force 0.089 -0.02 0.266 Dependent Variable = Reform Party Support. n=1840. Table 3.4 shows that Use of Force is the most important factor in the gender gap, as when it is removed, the significance of gender rises considerably. Feminism is more important than capitalism in creating the gender gap, but is still far less powerful than Use of Force. Capitalism is a significant factor in creating the gender gap, but the weakest of the three. R 2 will indicate the impact on Reform Party support, and shows that Capitalism and Feminism are -31-about equally important in determining Reform support, whereas Capitalism is a less significant factor. With this information we can answer two of our earlier questions. Men are actually more conservative than women, so we do not have to look to other possible causes of voting difference. While comparative mobilization strategies and the differing appeal of party leaders or local candidates may provide additional information, political ideology factors appear to be providing much of the difference causing the gender gap. Men's conservatism is not restricted to a single area of economic, cultural or authoritarian governance. Rather, men are more conservative in certain aspects of each of these general areas. Men are more pro-capitalism and pro-free markets, less supportive of feminism, and more likely to support the use of force to solve problems. However, these findings leave other questions unanswered. To find solutions to the questions posed we should be able to turn to the existing political science literature. But the approaches catalogued above have not addressed some of the political realities of the current gender gap. Which gender decides who forms the government? Why are men more conservative than women? Is there a male backlash against feminism? If so, what can be done to address the concerns of men and lessen the backlash's impact? These questions are not just matters of intellectual curiosity. With the rise of 'equality-as-sameness' ideologies in right-wing parties driven by male voters, the recent gains made by the women's movement are threatened, as are those made by aboriginals, minorities, and any other community calling for a collective sense of responsibility. A reconsideration of the existing analytical methodology is integral to comprehending the gender gap in its fullness. -32-Chapter 4 -Problems in the Literature: "It is men who face the biggest problems in the future, adjusting to their new and complicated role." Anna Ford This chapter will be a critique of the present bias in the literature exploring the gender gap that emphasises the contribution of a female voting block, and ignores the effect of men's voting behaviour. Popular beliefs that the gender gap is a product solely of women's political power and that the gender gap always benefits women will be exposed. The problems caused by focussing solely on one side of a dichotomous relationship will be explored, using three articles by prominent authors as examples. The factors that led to the present conception of the gender gap and the underlying theoretical structures of contemporary feminist theory will be discussed in order to identify the underlying causes of the focus on women in gender research. The chapter will conclude by considering the perceived unity of women as a voting block and establishing the basis for consideration of a corresponding male voting block. In research gendering voting behaviour, the tendency has been to place women as the independent variable to'be studied and men as a 'normal' control group against which comparisons are made. Carol Bacchi notes in her book Same Difference that: "It is, of course, possible to speak about the difference between one thing and another, but the usual meaning of difference is 'distance from a point of reference'. And, in Western conceptual systems, the feminine is always defined as a difference from a masculine norm" (1990: x). Masculinity, often confused with a dull neutrality, was forgotten as new attention was paid to the influence of gender on women in the aftermath of the Reagan revolution. While a focus on women has facilitated an understanding of the impact of gender on women's behaviour, another -33-effect is that male gender has been practically ignored. In Backlash. Susan Faludi writes that: "Unfortunately our social investigators have not tackled 'the man question' with one-tenth of the enterprise that they have always applied to 'the woman problem'. The works on masculinity would barely fill a bookshelf (Faludi, 1991: 61-62). The role of men in forging recent conservative victories has been overlooked by both traditional political science and feminist investigators. Men have been studied as a point of comparison, but not as active agents in electoral politics. Such an analytical framework can be observed in many of the key articles and books on gender and public opinion (Frankovic, 1982; Norris, 1986; Kopinak, 1987; Carroll, 1988; Erie & Rien, 1988; Sapiro, 1990; Cook & Wilcox, 1991; Rinehart, 1992; Gidengil, 1995; Everitt, 1996). Scholars'have investigated the effect of feminism on women voters, the effect of workforce participation on women voters, the effect of religion on women voters, but few authors considered men as more than the opposite side of the duality. Women are described as the gender that has experienced the primary shift in recent opinion, and much of the theoretical and analytical resources are aimed at explaining the behaviour of women as a group. This predisposition is especially obvious when one looks at more popular works (Rebick, 1998; Fraser, 1997). The impact of feminism coupled with women entering the paid workforce in large numbers have received enormous attention, and rightfully so. However, the political impact of this shift on men has received almost no investigation, because they are assumed to be static and uninteresting. The interpretation of data and conclusions focus almost solely on women and why they are different from men, rather than ever reversing the question. -34-Approaching a dichotomous relationship with an almost exclusive focus on one side produces a number of problems. First, the bias of investigation portrays men as the traditional 'norm' and women as outside the 'norm' when both groups differ from a gender-neutral simplified position. Traditional political science terms and concepts are assumed to incorporate men into them, while women are thought to be, because of their gender, different from these theorised subjects. Gidengil's 1995 article "Economic Man - Social Woman" is an example of scholarship that carves out a new conception of women as political actors, but leaves men trapped by traditional interpretations of political motivation. Employing the theory of Gilligan and other 'difference' theorists, Gidengil continues the charge that the rational choice conception of an individual as 'rational and calculating, utility-maximizing and non-altruistic' is a male construction. However, Gidengil herself argues that all voters seem to be moved more by 'sociotropic' than by 'egocentric' concerns, intimating that rational choice analysis may be overestimating the individual utility-maximization that goes into the voting decisions of both genders. While she concludes that "consistent with the image of economic man, men were indeed more likely than women to bring economic considerations to bear," there is no investigation of what effect gender has on these men (1995:403). Gidengil assumes that, while femininity effects the values and behaviour of women in ways heretofore rarely explored, the behaviour of men and the effects of masculinity have been perfectly captured by existing theories of the 'public man'. But, just as reproduction, sexuality, child care, and domestic labour have been relegated to the domain of the 'private woman', so to have aspects of male reproduction, sexuality and child care -35-been withheld from discussions of the 'public man.' The genderless 'rational actor' fails to describe accurately either women or men as women or men. Second, the focus on women disregards the possibility of substantive change in individual men and the political opinion of the aggregate group of'men' overtime. In her article "Sex Differences in Support for Government Spending," Deitch's own time-series data demonstrates that the movement of male opinion between 1976 and 1982 was more dramatic than women's, as can be seen in Figure 4.1 (1988). The scale measures support for government spending, with a high score indicating opposition to government spending. She demonstrates that both women and men increased in their opposition to public spending after 1974, but that men's opposition increased much more, producing a widening gender differential on this issue from 1975 to 1980. She notes that "the gender gap, however, decreased after 1980 because men moved closer to the women's positions, especially in 1983 and 1984." (1988:202) Figure 4.1: The Gap in Support for Government Spending in the United States (From Deitch) 9.5 9.4 9.3 9.2 9.1 9.0 8.9 8.8 8.7 8.6 8.5 8.4 8.3 8.2 8.1 8.0 Men Women 1973 1974 1975 1976 1977 1978 1980 1982 1983 1984 -36-However, Deitch's summary and conclusion does not address the fact that the gender gap in 1980 seems most attributable to Reagan's tremendous popularity among lower- and middle-income white men and the significant shift of these men to the right in the years leading up to the 1980 election (Goertzel, 1983). Deitch's explanation instead focuses on the changes in women's economic role since the 1960's. The dramatic shift in the aggregate attitudes of American men is downplayed greatly and the shift in attitude among women becomes the sole focus of her paper. Finally, the exclusive focus on women ignores evidence that demonstrates the influential role of male political behaviour. Wirls makes a convincing case that the traditional view of the gender gap has misidentified the source of that gap in the United States (1986). The gender gap observed in American elections in the early 1980's was a product of men's increasing identification with Reagan and the Republican party and "women have become the majority of the Democratic party by default and not by active realignment" (1986:327). Wirls concludes that the dominant interpretation of the gender gap - that it has been caused by the movement of women in the political sphere - has obscured the more accurate and important meaning behind the numbers; the dramatic increase of conservatism among men and relative stability of women in America has passed unnoticed. Wirls work was not the only investigation to observe that men were the cause of the 1980 gender gap, not women. Elshtain came to similar conclusions in 1984: "Women's votes in 1980 largely followed party lines and showed little change from 1976.... But men showed a significant shift to Reagan, and it was this change in male voting patterns that highlighted a gender gap.... One could make the argument that it was the men whose voting behaviour really should be explained" (1984:22). -37-Neither Elshtain nor Wirls' work is discussed substantially in subsequent research, despite the groundbreaking findings. Their conclusions contradict the dominant interpretation of gender as synonymous with women. If the gender gap in the 1980 election was the product of a male-dominated defection from the Democratic party, then the gender gap would not provide a vehicle for introducing women into political science research. The traditional explanations for the gender gap have produced strong insights into the effects of gender on women voters, but the corresponding effect of gender on male voters has been swept under the rug. There have been numerous books written by feminists that examine politics and masculinity (Bacchi, 1990;Enloe, 1990; Petersen, 1992; Tickner, 1992; Zalewski, 1998). But, all the works mentioned here study international relations, where men have historically dominated the decisions. Research on political men in domestic political settings is rare and almost never undertaken by 'feminist' writers (Ehrenreich, 1983;Mishkind, 1987;Pleck, 1987). Investigations that treat both men and women as active 'gender voters' are rarer still (Goertzel, 1983; Wearing and Wearing, 1991; Abramowitz, 1998). In-depth considerations of the political behaviour of men as men are almost nonexistent, with Daniel Wirls' article discussed above the only one I have encountered (1986). The lack of attention given to men by investigators of the gender gap results, in part, from a continuing lack of scholarship on gender from 'malestream' authors. It also stems from one of the principal critiques of traditional scholarship developed within feminist theory. Causes ofthe Focus on Women: A number of feminist writers have identified the human subject in traditional conceptions of political theory and the dominant political discourse as male -38-(Okin, 1979; Elshtain, 1981). Concealed behind this 'public man,' they argue, there exists a private world to which women are consigned through omission, tradition, overt practice and unchecked assumptions. "The public and the private as twin force fields help to create a moral environment for individuals, singly and in group; to dictate norms of appropriate or worthy action; to establish barriers to action, particularly in areas such as the taking of human life, regulation of sexual relations, promulgation of familial duties and obligations, and the arena of political responsibility" (Elshtain, 1981: 5). Those people who are assigned to the private sphere - wives, children, servants, slaves, - do not appear in the public sphere or in the theorising done about public life. Instead, men use the public sphere to keep these other groups in check (Pateman, 1988). Thus traditional conceptions of politics not only just represent a masculinized world, but also fail to address other forms of plurality and difference, imposing a suffocating uniformity on the human subject (Pateman and Brennan, 1979). "The term 'men' is used as an unmarked universal category to stand for humanity in general" (Cornwall, 1994:1). While this demonstrates an androcentric bias that has excluded the vast majority of humanity, the reciprocal effect has been a lack of thought about men as 'men'. If the term 'men' is employed to describe all of human kind, then it cannot be used to describe only those who are male. Harry Brod, in "Making a Case for Men's Studies," argues that "androcentric scholarship is only seemingly about men. In reality, it is, at best, only negatively about men, that is, it is about men only by virtue of not being about women"(l987:264). The traditional subject is certainly not a woman, exemplified by the absence of pregnancy and other female body characteristics and, as some feminists would argue, by the focus on a psychology of competitive atomisation rather than a psychology of nurturance. -39-But at the same time, the subject is devoid of any evidence of male sexualities and reproductive capacity, or any male role in child care or domestic labour (Carver, 1997). The traditionally conceptualised 'political man' was more of a 'political eunuch,' an unsexed neuter created in order to serve as a generalisation of humanity (Phillips, 1992). If 'public woman' is missing from conceptions of the subject in traditional western thought, so too is 'private man.' When the gender gap was first identified, the material was misjudged and this has had lasting implications in both popular and scholarly conceptions of engendered voting behaviour. In the days following the Reagan landslide, National Organization of Women (NOW) president Eleanor Smeal was devising a strategy to pursue during the final stages of the ratification of the Equal Rights Amendment when she came across a New York Times article by Adam Clymer discussing lower levels of support among women for Reagan (Clymer, 1980). "Nearly everyone looked at the election results as 'women: 45% Carter, 46% Reagan' - only a one percent difference, with fewer women voting for Carter. Smeal, the political scientist, saw in the data: 'Women 46% for Reagan, Men 54% for Reagan' - an 8% difference - a women's voting bloc" (Kenski, 1988: 85). Smeal recognised that gender was becoming a significant cleavage in society, and that women as a group were voting less Republican than men. Drawing on liberal feminist theory, the NOW president conceived of women as a category of people who share a set of experiences and interests that distinguish them from other [male] members of society and that these interests united women into a political unit. Drawing on both internal cohesion and great size, Smeal felt women could create a voting bloc more powerful than Catholics, union members, blacks or the South (Carabillo, 1993). Democratic consultant Betsy Dunn even coined a phrase to describe 'Reagan's female problem': the gender gap (Mueller, 1988). -40-As the 1984 election approached, NOW and the Democratic party expected the 'gender gap' to be a key factor in the defeat of Ronald Reagan. Top Democrats were convinced that if they nominated a woman for vice-president and made vague noises about women's rights, they could expect a wide majority of women to support them and regain the White House by exploiting the women's voting bloc (O'Neill, 1987; Bashevkin, 1998). Both Smeal and former congressional representative Bella Abzug wrote popular texts on gender-based difference in voting behaviour and political opinion that mapped out strategies for the Democrats (Smeal, 1984; Abzug, 1984). But the election did not unfold the way these leaders had hoped. The great oversight by NOW activists analysing voting results and the Democrats devising election strategy was that the 'gender gap' runs both ways. The common wisdom about gender-based voting divisions is that they will aid left-wing candidates who appeal to women. But gender-based differences in voting behaviour can be toxic to women of the left, especially if the nature of those divisions are as misunderstood prior to the election, as in 1984. Progressive observers failed to realise that men were operating as a simultaneous and opposing voting block of even greater strength. But the Republicans were thinking clearly on the issue. According to Reagan Campaign Director Ed Rollins, the goal in 1984 was not to eliminate the gender gap, but to identify women who would vote for the President in order to win a majority of women's votes (Kirchten, 1984.) Combined with their lead among men, this would ensure victory. What matters most for the gender gap in the United States is not how many more women are voting Democrat than men. What matters is if more women are voting Democrat than the men who are voting Republican. Only by being the more cohesive gender can women drive the gap. And in 1984, women were less cohesive than men. -41-The Democrats approach to the 1984 election is just one example of assumptions made that gender is a synonym for women. Rinehart contends that the dominant place of men in society, especially white men, means that men have no need to articulate their group's special point of view; it is the view of the system. She argues that while men may think of themselves as men, they have no group identity. And lacking identity, men cannot act cohesively as a group (Rinehart, 1992). But is subjective identification with a group necessary for that group to act coherently? Very few Canadians living outside of Quebec conceive of themselves as "English-Canadians," but we use this term constantly in discussions of electoral behaviour. Men may not be a completely unified block of voters to be mobilised in support of a particular candidate or party, but then again, neither are women. Subjective identification with a group identity, an articulate leadership and stated goals are not necessary for people to vote in a similar pattern. Common interest can remain unspoken. Each woman makes her own decisions about what policies or candidates will best serve the public interest and votes on the basis of individual decisions; the aggregate of all these decisions produces the 'women's vote' (Klein, 1985). While there is certainly a predisposition among women to vote in a certain way or to hold certain opinions, these differences are not as overwhelming as those based on race, ethnicity, or income in the U.S., or ethnicity or region in Canada (Burvis, 1984; Johnston et al, 1996). Women and "men do not appear to form two distinct groups, with separate and particular attitudes toward government and society, but rather two overlapping groups where agreement is more common than dispute. The key to the gender gap is to understand which group acts more unified than the other in making voting decisions. To speak of 'men' and 'women' as consistently unified groups is a synodic fallacy - the -42-tendency of observers to refer to a group or collectivity in terms of its central tendency, as if it had no internal variation at all (Bacchi, 1986). Men have a gender. To 'add women and stir' is not all there is to gender politics. Rather, men's gender is central to understanding why this large and powerful group behaves the way it does. Gender is integral to how men conceive of the world. It may affect different men in different ways in different contexts. It helps to explain the root causes of war (Wiener in Strauss, 1998) or why men must always have the television remote control. According to Carver: "Moving private man into the political foreground reinforces the feminist critique of masculine violence, domination, and privilege by looking at what men fail to do altogether, what they claim to do symbolically rather than carry out practically, and what role their sexualities play in their collective and individual constructions of power" (Carver, 1996:4). The ideological, situational, structural and socialised effects of gender on male voters has been an oversight in the investigation of the gender gap; an oversight that leaves many progressive parties unable to address their current deficit among male voters, while providing conservative parties the tools to attract more women1. The focus of gender gap investigations must broaden if they are to address the entire spectrum of voters and be of use to progressive parties. i The most dramatic use of research on the gender gap in an election was George Bush's victory over Michael Dukakis in the 1988 American Presidential election. Bush adopted moderate positions on issues like the environment while stressing his tough positions on crime and patriotism, a communitarian strategy designed to turn around his twenty point deficit among women. In the end he won a majority of women's (and a larger majority of men's) votes (Jamieson, 1995). -43-Chapter 5 - Voting Blocks and the Gender Gap "Man is the measure of all things." Protagoras (c. 485 - 410 B.C.E.) While Protagoras' statement demonstrates the androcentrism of early scholarship, this chapter will serve to demonstrate that 'men' do have a significant role in the outcome of elections. In order to determine the state of the gender gap and the relative power of each gender, we will be drawing on rational choice theory of pivots and power in coalitions as developed by Arrow (1951), Shapley (1953), Downs (1957), Riker (1962), Rae and Taylor (1970), Axelrod (1972), Johnston et al (1992), and Bakvis and MacPherson (1995). Five factors key to power in voting blocks will be considered: size, turnout, elasticity of response, positioning and cohesion. Once this theory has been established, we will be able to operationalise it to test the hypothesis that men are an integral part of the gender gap in Canada. However, the Canadian multi-party system makes discussions of the relative power of voting blocks very complex because of the multiple relationships amongst all the parties. Simple examples will be employed to illustrate the theory before engaging our principal objective. Periodically, non-gender examples will be drawn from American and Canadian politics to illustrate points throughout the text. Before exploring the Canadian system, elections to the United States House of Representatives will be analysed as a simple two-party system with an important and mobile gender gap. Following the U.S. House example, the Canadian gender gap will be tested to determine the relative power of each gender in government formation. Several -44-models will be considered to determine the relative importance of each gender during different periods in Canadian politics, and between parties and 'families of parties.' The final section of this chapter will explore the targeting of male voters in elections. The women's movement and research drawing on an orthodox conception of the gender gap have developed a wealth of information about women voters and the issues that mobilize, attract and repulse that group. However, little conscious thought is given to targeting male voters. This section will identify issues subjectively identified as key in the 1997 election by male and female voters. While incomplete, this discussion helps, in part, to begin the process of treating men as a serious voting block. Reconceiving the Gender Gap: Coalition analysis of voting blocks involves five factors: size, turn-out, positioning, elasticity and cohesion. Size matters in that a larger group can contribute more votes to a party than a small group can (Axelrod, 1972). This fact should not be underestimated. Minor opinion movements in a large group like Catholics or women will produce a larger net shift in votes than major moves in a smaller group like homosexuals or students. Gender, under the definition developed in the introduction, divides the population into two large groups, each incorporating approximately fifty per cent of the population. The sheer size of the groups 'men' and 'women' means that relatively minor differences in opinion translate into large differences in votes measured in the hundreds of thousands. It is due to size that gender, or any voting block, matters. Turnout is a factor related to size in the population and affects the power of a group by strengthening or lessening its stature in the electorate. An efficient and mobilized block with a -45-high turnout can wield more political power than a much larger group with a very low turnout (Axelrod, 1972). Short of encouraging more births or introducing selective immigration, groups cannot increase their size. However, mobilization and registration of voters can improve turnout, thereby increasing political power. Turnout is an important factor in American elections, where voting levels are quite low. Women's turnout was higher in each presidential or mid-term election from 1980 until 1998, when a greater percentage of men voted (Seltzer et al, 1997; Berke, 1998:9). In Canada, women were only slightly less likely to vote in the 1997 federal election than men, according to the CNES. However, turnout is a trivial factor in Canadian elections (Eagles, 1991; Black, 1991). Canadian turnout levels are much higher and more stable than the US, leaving fewer voters for mobilisation efforts to target. The effect of positioning is derived from the electoral system. A by-product of the Single Member District Plurality system used in Canada is that elections are not won by votes, but by seats. As a result, parties do not expend resources in regions where support is guaranteed or impossible. Parties instead target winnable ridings where money, policies and patronage may swing the handful of votes crucial to winning those ridings. Platforms will be tailored to attract supporters in those areas where slight increases in votes result in meaningful changes. If a voting block is present in numerous critical ridings, where a swing of relatively few votes will produce a large net gain in seats, that group will be able to demand greater policy rewards. Even if such a voting block remains outside the winning coalition, it may draw government payments if its positioning appears critical in an upcoming election. Voting blocks that are in a position to shift -46-a number of seats will attract a greater number of policy rewards than a voting block that is not as fortunately arrayed. Elasticity is linked to political leaders' perceptions of whether a group is worth courting. It is the political equivalent of flirting: a harmless way to keep party leaders interested in a potential future liaison. Elasticity increases the political power of a voting block by demonstrating the block's receptiveness to appeals for support and its dramatic reaction to greater or lesser attention. Axelrod calls this a group's "elasticity of response," where greater attention is paid with greater support and lessened attention results in diminished support (1972:19). Elasticity measures the mobility of a voting block in response to policy promises and action by government and opposition, and is indicated by the standard deviation. Parties that register a high standard deviation will be have greater fluctuations between results in a series of elections. Groups that have been targeted by parties, but fail to move their support have lower levels of elasticity than groups that are highly mobile. However, sweeping generalisations about elasticity should be avoided, as targeted appeals may be muted by loyalty, distrust and poor messaging. Due to their volatility, highly elastic blocks often occupy the pivotal position of the last-added members of a winning coalition (Axelrod, 1972.) The pivot is crucial to victory and thus receives the highest prize in policy and patronage (Riker, 1962). A group with known elasticity will usually attract the favour of both governments and opposition groups as their pivotal position makes them key to victory. A block with high mobility between parties will draw greater rewards than a block that is solid in its support of a single party ceteris paribus. Great -47-elasticity brings with it a central role in deciding questions of government formation, and a powerful place in the post-election distribution of goods and policy. American Southern whites are an example of a group whose elasticity made them the pivotal voting block in their national electorate. The nomination of Texan Lyndon Johnson in 1964 raised the Southern vote for the Democrats by 6 points more than 1960, despite the passage of the Civil Rights Act that year. When the attention of the Democrats shifted to the Northern states in 1968 and Minnesotan Hubert Humphrey was nominated, the Democratic vote collapsed by 19 per cent in the South (Axelrod, 1972). The regional campaign of former Alabama governor George Wallace and Nixon's Southern Strategy helped to draw voters from the former Confederacy away from their traditional home. By 1972, the South had moved to the Republican camp. Since then, the Democrats have only been able to win the presidency when they nominate Southerners like Carter and Clinton and can attract the pivotal block of white Southerners. The fifth factor in determining the power of a voting block is its tendency to behave like - a block, or its cohesion. Bakvis and MacPherson define cohesion as "the propensity of a specific group or region to give concentrated or disproportionate support to one or another party in any given election relative to support by groups or regions for the largest party elsewhere" (1995:664). A group that votes overwhelmingly for a party will be responsible for a larger than proportionate share of that party's victory. When comparing groups which, like genders, are of relatively equal size and turnout, the pivotal group will be that which is most biassed toward one party. Wirls warns that "no Republican need fear, nor can any Democrat take comfort from, a gender gap wherein women give a smaller majority of their votes to the GOP and a bigger -48-minority of their vote to the Democrats than their male counterparts" (1986:324). The greater cohesion of men to the Republican party in the 1980's won them the White House and demonstrated that men were a voting block in elections. The tendency of a group to vote for a single party in a multi-party system cannot be expressed in absolute terms, but may be compared relative to another group. The fact that 23 percent of men voted for the Reform party is, on its own, no measure of cohesiveness. Also needed is the knowledge that 15.7 percent of women voted for the Reform party. So measures of cohesion must be made in a comparative setting between groups. Cohesion is best expressed as the distance from the threshold of victory. If one gender is cohesive enough to produce victory despite a lack of matching support within the other gender, the support of the winning gender will be further above the threshold of victory than the opposing gender is below. The gender gap is a poor attempt to explain cohesion in a two-party, two-gender system. The fact that 9 percent fewer women than men voted for Reagan is a meaningless statistic. What is pertinent is the fact that while women were tied in their opinion of the two presidential candidates, nearly twenty percent more men voted for Reagan than Carter. When considering the relative importance of each gender in questions of government formation, the key variables are elasticity and cohesion. Size is irrelevant because the genders are basically of the same stature. Turnout is also not a factor in relative power between the genders. Even in the United States, where turnout is low and political mobilization is an important factor in electoral victory, the efforts of women's groups to register a greater number -49-of women have been only barely successful1. Entire genders cannot be thought of in terms of positioning, as each sex is arrayed across the nation relatively evenly. Only subgroups of each gender - such as feminists, anti-feminist women, young men, gay women - are found in geographic clusters, and these subgroups are not the focus of this study. Also, the importance of a group's position changes over time, as the pivotal 'battlefield' ridings that will decide an election shift. Our analysis will not be sensitive to positioning factors. Elasticity and cohesion are the key factors in comparing the relative power of voting blocks like gender; blocks that are of equal size, turnout and positioning. The most powerful voting blocks possess both elasticity of response and cohesion when they move. A voting block that is responsive, but unable to act cohesively will attract some attention from political leaders. However, because the group's cohesion is low, attention will not produce large shifts in votes. A voting block that is stable in its support of a single political party and resists attempts to shift it to a new party will gain patronage and policy benefits from only the party that it hews to. Other parties will have only limited interest in that group. The most powerful voting blocks are both elastic and cohesive, shifting overwhelmingly to whatever party will do more to address their concerns. These groups will be the focus of attention from many parties, and addressing their concerns will be a focal point of debate. The province of Quebec is an example of a cohesive and elastic voting block. John McMenemy calls Quebec "a province and a region that has done much to decide the outcome of virtually every national election since Confederation" (1976:14). Before 1984, Quebeckers i Despite the best efforts of American feminist groups, US women's turnout has never surpassed 52% and fell below 50% in the 1998 mid-term elections (Seltzer et al, 1997; Berke, 1998). -50-supported the Liberal Party of Canada overwhelmingly, providing the Liberals with one quarter of the seats in parliament each election. The Liberals then only had to win one in every three English Canadian seats to form a government. However, Cairns demonstrates that Quebec's one-party monopoly is only at the level of seats, not votes (1968). Only slightly more than half of Quebeckers voted for the Liberals, but enough to provide the maximum number of seats. The NDP and Tories could not ignore Quebec because of its size, and its positioning and elasticity presented a situation where meaningful attention would be rewarded with seats. In 1984, Quebec switched its allegiance to the Conservatives, acting as the pivot and were the last-added members of a winning coalition. Quebec was perceived to be the crucial element in victory for any party, and by bringing Quebec on side, Brian Mulroney was credited with creating a durable Progressive Conservative winning coalition. Quebec shifted overwhelmingly in terms of seats, but remained available in terms of votes (Johnston et al, 1992; Bakvis and MacPherson, 1995). By remaining responsive to serious attention from various parties and shifting en bloc to the best deal, Quebec has wisely used the electoral system "as an instrument of racial defence, or of personal gain" (Trudeau, 1968). However, by movign to the Bloc Quebecois and removing themselves from the government formation game, Quebec has lost much of its power as an electoral bloc. Testing the Framework: Looking at the gender gap as a system of voting blocks, it is best to illustrate the logic on a simple two party system before engaging the complex Canadian multiparty order. The United States provides several examples we can use, but elections to the House of Representatives are the simplest, provide a high number of cases and are most like -51-parliamentary elections. With 435 elections every two years, short term fluctuations caused by individual candidates and campaigns are kept to a minimum in the national results and more fundamental questions of partisan loyalty are visible. Table 5.1 measures the cohesion for each gender across time, defined as the distance from 50 per cent, the approximate total needed to control the House. The gender that is further away from 50 per cent is demonstrating greater cohesion than the other. If 42 per cent of men support the Democrats, then 58 per cent support the Republicans and, if only 53 per cent of women are supporting the Democrats, the more cohesive male block will win the election for the Republicans. Table 5.1: Gender Gap in U.S. House Election 1978-1996 (Percentage Voting Democrat) Year Male Vote Female Vote Gap ( ¥ - 0 * ) Cohesio n Male Cohesion Female 1978 59 58 -1 9 8 1980 49 55 6 1 5 1982 55 58 3 5 8 1984 48 54 6 2 4 1986 51 54 3 1 4 1988 52 57 5 2 7 1990 52 55 3 2 5 1992 52 55 3 2 5 1994 42 53 11 8 3 1996 45 54 9 5 4 Column Mean 50.5 55.3 4.8 3.7 5.3 Column o 4.8 1.8 Initial Data from Seltzer et al, 1997 and Whirls, 1986. Cohesion = absolute difference from 50%. -52-In 1978, the traditional gender gap was in its final days, with more men supporting the Democrats than women, and both genders overwhelming in their support of Democratic candidates. In 1980, the modern gap emerges. Men shifted dramatically to the right in their choices for the House, due to Reagan's coattails. In doing so, men demonstrated tremendous elasticity, responding to Reagan's message of laissez faire economics and hardline foreign policy. However, Democrats held the House because women remained relatively unchanged, as they did throughout the period in question, shifting only three per cent. In 1982, men returned to the Democrats briefly, only to shift even more strongly behind the Republicans in 1984. Between "Morning in America" and "Contract with America" the gender gap stabilized, with both sexes supporting the Democrats and women more so. Women were the more cohesive group before 1994, and were able to counterbalance an increasingly conservative male population. However, women displayed little elasticity during this period. The nature of the gender gap fundamentally altered in 1994. Men became the more cohesive group, overwhelming the majority of women still supporting the Democrats and electing the first Republican House of Representatives in fifty years. Abramowitz and Saunders demonstrate that the movement and cohesion of white males contributed overwhelmingly to the Republican takeover of congress and the realignment in the United States, proving the existence of a white male voting block and demonstrating conclusively that it had the power to win elections (1998). Men had displayed a greater elasticity during the 1980's, moving between the Republicans and Democrats several times. The standard deviation of the mean for men was two-and-a-half times that of women, who remained consistent in their support across the period. The early gender gap was caused by men moving to more conservative positions, although men were -53-not cohesive enough to overcome women's continuing support for the Democrats. At present, men are the more relatively powerful voting block, with greater cohesion and elasticity than women. Figure 5.1 Gender Gap in U.S . House Elections 1978-1996 (Percentage Voting Democrat) 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 Initial Data from Seltzer et al, 1997 and Whirls, 1986. Cohesion = absolute difference from 50%. Figure 5.1 makes the story abundantly clear. Women have slowly been leaving the Democratic party over the past twenty years, although a majority of women still vote for the Democrats2. Men left the Democratic party in two large shifts. In 1980, the first shift made the group 'men' basically party neutral, although they were highly volatile during this period. The The overall formation of gender cohesion has changed in concert with the recent shift in which party controls which branch of the American government. Before 1992, women were cohesive enough to win House elections for the Democrats, but men won the presidency for the Republicans. In 1992-1994, that arraignment inverted, with men cohesive enough to throw the Congress to the Republicans, but women rallying to elect, and then reelect Bill Clinton as president. -54-late 1980's saw men's votes stabilize at about 52 per cent support for the Democrats before another massive swing brought them over to the Republicans in 1994. Men have displayed greater mobility and elasticity during this period. Women have been more stable in their support of the Democrats. Only Reagan in 1980,1984 and 1986 was able to draw women away from their usual pattern of Democratic support. Applying the Framework to Canada: In the United States, a two-party system of government means that 50 per cent of the vote over all the elections to the House of Representatives is the approximate threshold required to control Congress. However, Canada is a much more complex party system. While Canada also employs a First-Past-The-Post Plurality electoral system, its multi-party dynamic can produce majority governments for a single party with far fewer ballots. The 1997 Canadian federal election resulted in a majority Liberal government of 154 seats, even though the Liberals received only 38 per cent of the vote. In contemporary Canada, 50 per cent of the vote for a single party would result in a landslide majority government and a rump opposition. In order to determine the threshold for victory in Canada, we computed the vote to seat ratio for the modern election period and used that to determine the votes typically required to secure a majority in the House of Commons. The absolute value required to produce a majority government shifts in each election, so we determined its average value across the modern election period and measured cohesion from that figure. The following formula was used: X = (Y- a) + 0 -55-In this equation, X equals the vote share required to produce 50 per cent of the seats, Y equals the number of seats (in this case 50), a is the intercept of the relationship between votes and seats in Canadian election results, and ft is the slope for the period. The data in question used all federal election results of the Liberal Party, Progressive Conservative Party and New Democratic Party from 1935 to 1993 to determine the intercept and slope. Calculating the average vote to seat relationship of the three major parties from 193 5 to 1993 produced an intercept of-18.71 and a slope of 1.652. Solving for 50 seats in the House of Commons, our model above predicted that 41.59 per cent was the approximate threshold required to produce a majority government. X = 5 0 - (-18.71) H- 1.652 X = 41.59 Since a variety of governments have been formed with a smaller total than 41.59 per cent of the vote, we rounded down to 41 per cent.3 Arguably, Canada entered a new party system in 1993 in which an even lower total vote is required to form a government.4 To test for any effect of the new era, the 38 per cent vote that allowed the Liberals to barely retain their majority in 1997 was also tested and no major differences between the two were found. 41 per cent is the average threshold required for majority victory in a Canadian federal election and measures of cohesion in the remainder of this paper are from that level. 3 While Joe Clark formed his government in 1979 with only 36% support, that was under extraordinary circumstances. 4 Bob Rae and the NDP in Ontario (37.5% in 1990), Glen Clark and the NDP in BC (38% in 1996) and Jean Chretien and the federal Liberals (38% in 1997) are recent examples of majority governments formed with a bare plurality of the vote. -56-The problems of over-reporting from the pre-1988 Canadian National Election Studies further complicate the picture in Canada. As in the study of the gender gap in Chapter Two, the data must be refined to raise those parties who were under-reported and lower those that were over-reported. The same formula will be used as in Chapter Two. Please consult Appendix One to compare the refined and unrefined data, the methodology and reasoning. Our first table will calculate cohesion based on the largest vote for a single party in each election by each gender. The votes of a gender block will be concentrated if the total is high and dispersed if the total is low. Tests of elasticity will not be available on this table, as the figures vary between parties and elasticity is a measure of the variance in support of a single party. Table 5.2: Electoral Cohesiveness - Mean Percentage Vote for the Largest Party in Canada 1974-93 and 1984-1993 (Refined Data) Year Winning Party (Maj ority/Minority) Largest Vote Male (Party) Largest Vote Female (Party) Cohesio n Male Cohesion Female 1974 Liberal (Maj) 38.6 (L) 47.4 (L) 2.4 6.4 1979 Conservative (Min) 40.7 (C) 43.9 (L)* 0.3 2.9 1980 Liberal (Maj) 39.5 (L) 48.6 (L) 1.5 7.6 1984 Conservative (Maj) 51.6 (C) 48.5 (C) 10.6 7.5 1988 Conservative (Maj) 46.8 (C) 39.2 (C) 5.8 1.8 1993 Liberal (Maj) 39.1 (L) 42.9 (L) 1.9 1.9 1997 Liberal (Maj) 37.1 (L) 37.4 (L) 3.9 3.6 Mean 1974-1997 41.9 44 1 3.7 I 1 4.5 Mean 1984-1997 43.6 42 5.5 3.7 Taken from Canadian National Election Studies. Cohesion = absolute distance from 41%. *Not election winner. -57-The refined data for Canadian elections between 1974-1997 shows that women were the slightly more cohesive group through this entire period. As with the United States, women were more one-sided in their support of a single political party, and propelled the Liberals to majority victories under Trudeau and Chretien. However, solid support for the Liberals in 1979 did not prevent the Conservatives under Clark from forming a minority government. This may be understood that women denied Clark his majority, or that the wishes of a disproportionately male group of voters defeated a more cohesive female-dominated vote through efficient electoral distribution and positioning. However, the cohesion of women to the Liberals is restricted primarily to the Trudeau years. Women shifted massively to the Conservatives in 1984, and almost exceeded their support for the Liberals in the previous election. Men remained with the Conservatives in 1988, and, added to the bare plurality victory for the Tories among women voters, produced a second majority. Both women and men swung to the Liberals in 1993, although women made up a smaller majority of the victorious coalition than in earlier years. Female cohesion to the Liberals disintegrated by 1997, although male support remained at levels comparable to the Trudeau period. The story changes if we remove the Trudeau era from the statistics. Through the four elections 1984-1997, men are more cohesive in their support of the winner with an average cohesion level of 5.5 compared to women's 3.7. The 1984 election was gender neutral. 1997 saw only two minor parties with significant voting differences between the genders, the New Democrats and the Reform Party. 1988 saw men much more cohesive in their support of the Tories, and women split 35 per cent Liberal - 39 per cent Conservative. Had only women voted in the 1988 election, it is unlikely the Free Trade Agreement would have passed as a minority -58-government would have been the result. In 1993 women were more supportive of the Liberals than men. But it is interesting to note that while the support among men for the Liberals in 1993 is almost identical to support given in 1974 and 1980, five per cent fewer women voters supported the Liberals in 1993 than in earlier victories. The Liberals retained a greater fraction of their male supporters in the 1990's than female supporters. While this table does give us insight into the general cohesion of each gender, it does not measure elasticity across time. Table 5.3 has been calculated to determine the continuous level of support for the part(y/ies) most dominated by each gender. For women voters, this is the Liberal Party. However, men were predisposed toward the Conservatives before 1993 and Reform after 1993. The current split among conservative voters makes it difficult to compare across the two periods. But if we are to explore the effect of gender on voting behaviour during this period, such comparisons are inevitable. Table 5.3 will follow the support levels of the conservative parties throughout the last seven elections, measuring the combined support of both the Progressive Conservative and Reform parties after 1993. The Bloc Quebecois will not be considered part of the traditional conservative coalition because of the temporary nature of their addition to the Tory coalition in 1984 and 1988, and the evidence provided by various scholars that the Bloc attracted a large number of previous nonvoters (Johnston et al, 1995; Blake and Erickson, 1998). A comparison of the combined Tory-Reform vote to that of the Liberals also allows supporters of the United Alternative to continue their sugar-plum dreams of an eventual victory5. 5 Surveys during the 1997 election found that more Progressive Conservative Party supporters chose the Liberals as their second choice than the Reform Party, undermining the concept of a united conservative faction under Reform leadership (Angus Reid, 1997). -59-The NDP became significantly more female supported in the 1997 election, sparking consideration of the merits of studying the combined NDP/Liberal left-of-centre vote. But the NDP will not be included in consideration of women's traditional voting for two reasons. First, because this is a new phenomenon. For most of their history, the NDP had a net deficit among women voters and it is only in the most recent election that significantly higher numbers of women voted NDP than men. Second, the NDP is not a former faction of the Liberals, nor are New Democrats and Liberals taking of merging their parties. There are obvious reasons for considering the combined Reform/Tory vote, and none for the two progressive parties. Table 5.3: Electoral Mobility - Mean Percentage Vote for Gender Dominated Party Groupings 1974-93 (Refined Data) Year Largest Male Dominated Party Male Vote Male Cohesio n Largest Female Dominated Party Female Vote Female Cohesio n 1974 Conservative 37.5 3.5 Liberal 47.4 6.4 1979 Conservative 40.7 0.3 Liberal 43.9 2.9 1980 Conservative 38.9 2.1 Liberal 48.6 7.6 1984 Conservative 51.6 10.6 Liberal 29.7 11.3 1988 Conservative 46.8 5.8 Liberal 35.3 5.7 1993 Reform + Tory 37.3 3.7 Liberal 42.9 1.9 1997 Reform + Tory 40.3 0.7 Liberal 37.4 3.6 Column Mean Male Cohesion 41.9 3.8 Female Cohesion 40.7 5.6 Column a 3.2 3.2 Taken from Canadian National Election Study. Cohesion = absolute distance from 41%. Data refined. See App. 1. The average vote for conservative parties among men is higher than that of women for the Liberals. Men's support for the conservative parties has never been as low as women's -60-support for the Liberals under John Turner. However, women demonstrated greater cohesion across this period. Their vote levels were further away from the threshold of victory, indicating a greater tendency to shift en bloc. Both men and women had identical levels of elasticity, although the range of support for the Liberals among women was much wider. Men cohesively backed the Conservatives only twice over the seven elections in question, both during the Mulroney era. During the all the other elections, men's support for conservative parties has been just under the level required to form a majority government. In contrast, in the majority of elections in question, women supported the Liberal Party at levels predicted to produce a majority government. Canadian women have been more cohesive in their support of a single political party, and have shifted further away from that party than men have from their typical party to support when the Liberals acted in ways that displeased women. Women demonstrated elasticity of response in 1984, when Turner's behaviour and neo-liberal rhetoric repulsed women. However, in 1988, women were not able to coalesce behind a single party in opposition to the Free Trade Agreement. Despite the gender gap in support of liberalised trade in 1988, a larger group of women supported the Conservatives than the Liberals (Gidengil, 1995). The Liberals were able to win back most of their female supporters in the 1993 election when they ran as welfare Liberals and promised Keynesian job creation and child care programs. However, four years later their support among women was 9 per cent lower than during the Trudeau years. Figure 5.2 illustrates level of support for the Liberal Party among women across time. During the Trudeau era, women supported the Liberals at levels high enough to produce majority victories, despite some reluctance among men. During the Mulroney years, this support collapsed, although there were significant signs of recovery in the -61-1988 election. In 1993, Liberal support among women increased, but not to the levels experienced under Trudeau. In 1997, support for the Liberals among women was only marginally higher than it was in 1988. Figure 5.2: Electoral Cohesion - Mean Percentage Vote for Gender Dominated Party Groupings 1974-93 (Refined Data) 54 1974 1979 1980 1984 1988 1993 1997 Source: See Table 5.3 Women have left the Liberal party, presumably over the movement to business liberalism, the reduction of health, welfare and education spending, and the cancellation of key projects like nationalised child care (Barlow, 1995; Sharma, 1997). The Liberals should be concerned, as extra support among women has often been the margin between minority and majority government. Enacting legislation that will renew their privileged relationship with women voters should be the Liberal Party's priority in the remainder of this mandate. Women have proven their elasticity in past Canadian elections and direct attention to widely perceived -62-needs will likely be rewarded with renewed support. If women continue to exit the Liberal Party for the NDP, majority governments could become a historical oddity in Canada. In comparison, men have never left the conservative family of parties. In elections where the Liberals scored large majority victories, men's support for the losing conservatives was just below the minimum requirement to form a majority government in a multi-party single-member-plurality electoral system. Figure 5.2 shows that while support for the Conservative Party was only decisive in 1984 and 1988, men continue to keep the conservative parties viable. Male support for conservative representation hovers consistently between 38 and 41 per cent. Men's elasticity seems less driven by policy than simple lack of concentration. If men are disciplined and organise behind a single party, as they did in 1988, their ideals will be implemented (Wearing and Wearing, 1991). Today though, men are impotently aligned in two conservative movements. The key factor in contemporary elections is the split in the right. Men were finally able to coalesce behind a conservative party in substantial numbers in 1984. In 1988, men remained with the Conservatives. Men were cohesive enough in their support to overcome women's opposition to liberalised trade and produce the majority victory that Mulroney needed to pass the Free Trade Agreement. However, between 1988 and 1993, a large number of men (and significantly smaller number of women) left the Conservative party for the Reform party6. By dividing their support between two parties, conservative men are unable to implement their agenda. Women are showing signs of becoming as undisciplined as men are, recently shifting 6 Kim Campbell's only lasting contribution to the Progressive Conservative Party may have been the retention of enough women voters in the 1993 election to keep the party from collapsing completely. -63-support to the New Democrats. If progressive women also divide their support between two parties, gender will become a non-factor in questions of government formation. This series of tests has demonstrated the important role than men play in elections in North America. American men are the driving force behind the current Republican majorities in the House and Senate, and were responsible for the election of Ronald Reagan in 1980. While men are not responsible for the current Canadian government, neither are women, as the gender gap in Liberal support has decreased. But men were integral to Mulroney's victory in 1988 and the passage of the Free Trade Agreement. Men's predispositions toward conservative parties are keeping the Conservatives and Reform viable. If the 'unite the right' movement is successful and the Liberals do not act to shore up their support among women, a male dominated coalition of conservatives will contest a gender neutral Liberal party for government formation. These data is not meant to give the impression that women are not important in considerations of gender difference in voting. Men should no more be the single centre of attention than women should be. This analysis is meant to demonstrate that both genders matter when conceptualising the gender gap and that the current deficit in thinking about how males' gender affect them leaves us unable to explain the present rightward shift in voting, most visible at present in the United States. The growing conservatism of white men should be a major concern to feminist, minority and progressive political figures and scholars constructing strategies to implement new policies. White men retain a huge surplus of power to this day and understanding the motivations and values of men may help to mitigate the potential for conflict. -64-Targeting Men: If men are as much a cohesive voting block as women are, then parties should devote resources to attracting male supporters. Those parties that have had only limited success with men, like contemporary New Democrats, may want to consider advancing issues that would increase support among men. It is common knowledge that issues of health and education can be used to target female voters, but what issues can be used to target male voters? As one campaign manager noted in Fox's study: "There is a clear sense of what women's issues are. It is easy to put together a package of things for women only. It doesn't work that way for men" (1998:46). This section will identify issues potentially important to men and those championed by women. Table 5.4 tests a series of issue clusters from the 1997 CNES to see what percentage chose that issue as the most important issue of the campaign. The results are divided by gender and tested for significance. Those topics that produced a significant gender division are grouped into men's issues and women's issues, with insignificant results in a third category of gender neutral issues. The tests employed a series of two-variable regressions, with an isolated answer or cluster of answers for Most Important Issue as the dependent variable and Gender as the independent variable. The clusters of answers for Most Important Issue were developed using coding categories from the technical documentation for the 1997 Canadian Election Study (Northrup, 1998). However, these data was not cross-referenced with party support due to space and time limitations, and so has only limited analytical application. Issues that are gender neutral include the most commonly mentioned issue: jobs. Both groups, men and women, mentioned jobs more often than anything else as their principal concern during the election period. Jobs was twice as common as the next most important issue -65-among men (Quebec/Unity), and three times as common as the next most important issue among women (Education). Taxes, Pensions and the Environment were other issues tested that proved to have no significant gender division in importance in 1997. Table 5.4: Most Important Issue in 1997 Election Divided by Gender Most Important Issue Female Male Difference Gender Neutral Issues Jobs 31.2 29.1 2.1 Taxes 4.3 5.5 -1.2 Pensions/Elderly 1.5 1.0 0.5 Environment 0.6 0.8 -0.2 Issues More Important to Women Health Care 10.4 5.4 5.0*** Education 11.6 7.9 3 7*** Social Spending/Welfare 2.9 1.0 2 9 * * * Minority Issues 1.0 0.1 0 9 * * Issues More Important to Men Quebec/Unity 7.5 14.6 _7 1*** Finances/Deficit/Debt 5.7 11.2 _5 5 * * * Representation/Integrity 3.5 6.3 -2.8** Economy 2.4 4.8 -2.4** Crime/Gun Control 1.4 2.8 -1.4* *** p > .001; ** p > .01, * p > .05. Using variable cpsal. Data from CNES. The four issues that were of importance more to women than to men were Health Care, Education, Social Welfare Spending and Minority Issues. Women were more than twice as likely to name Health Care as an important issue than men, and almost two-thirds more likely - 6 6 -to name Education than men. Women were almost three times as likely to name Welfare and Social Spending. While one percent of women named Minority Issues as the key to the election, one-tenth as many men agreed. The difference over 'women's issues' helps to explain the gender gap in support for the New Democratic Party. Health Care, Education and Social Spending were issues at the centre of the 1997 New Democratic campaign (NDP, 1997). As the Liberals move to the right on economic and social policy questions, women with concerns about these issues are increasingly turning to the New Democrats. Issues that significantly more men than women found salient were a surprising lot. Some feminist scholars argue that the issues that dominate debate are 'men's issues' and to a i certain extent, this test has proven that true. Quebec/Unity is mentioned by twice as many men as women. Finance/Deficit/Debt and the Economy were identified as the key issue in the election by double the number of men than women. These two issues, along with the 'motherhood' issue of jobs, have dominated the agenda in Ottawa for the past decade. But two issues appear in the male camp that were unexpected and differ from the dominant issues of the day. Representation/Integrity was cited by twice as many men as a key issue in the 1997 election. These were issues about who would form the government, how that government should make its decisions and to whom representatives should be responsible. The second issue was crime, including gun control. It was not surprising to see more men than women cite the Firearms Bill (C-68) as a key election issue, with men making up the vast majority of the gun owners, hunters and sportsmen who were affected by the legislation. But the male preoccupation with justice extended to other questions of crime, such as harsher penalties for criminals and young offenders, and general mentions of crime and violence. -67-Overall, men were twice as likely to cite crime and punishment as the major issue of the election. The gender gap in support for the Reform Party can be traced in part to issues stressed by the organisation's campaign. The decision to emphasize issues that have greater appeal to one gender is likely not the product of a conscious decision, but rather the outcome of the general philosophical and ideological perspectives of the parties' leaders and members.7 The Reform Party's emphasis on the issues of crime and punishment helps to account for the gap in opinion on the issue of Use of Force. If disproportionate support for tough measures on crime are generalised among the male population, then men would be more attracted to a party that espouses a harsh, 'get-tough' rhetoric about crime. Similarly, by stressing issues that are more important to men than women, like the deficit and debt, representation, and unity, Reform attracts an androcentric support base. More advanced inquiries of the relationship between men and political issues should be undertaken, as there do appear to be some positions that will attract male voters. 7 Looking at two NDP and Reform Party brochures from the 1997 election, the gender gap in their core support is obvious. Three pictures of smiling Reform Youth from their brochure imaginatively entitled "Reform Youth" display a total of 32 men and 10 women (Reform, 1997b). A crowd of youth supporters from the cover of the "Highlights from the NDP Platform" show 10 applauding women and 5 men (NDP, 1997). -68-Chapter 6 - Discussion and Conclusions: "The male sex still constitute in many ways the most obstinate vested interest one can find." Lord Longford The previous chapter demonstrated that men are as significant a voting block as women. In light of evidence that men behave as a block, attention should be paid to the characteristics of male voting behaviour and its causes. Unfortunately, there has not been extensive research into the reasons for men's behaviour. Traditionally, political science explanations of the underlying factors contributing to voting difference focus on why women are more progressive than men on economic issues and some social issues. This chapter will invert that traditional approach and offer potential explanations for the gender gap based on male political behaviour. The entire story of the gender gap cannot be told in this way, but interesting angles that have been ignored may be discovered. This chapter will explore the roots of men's greater economic conservatism, militarism and opposition to feminism in the context of their disproportionate support for the Reform Party of Canada. Men's predisposition toward support for forceful solutions to domestic and international problems will be discussed with elements of socialisation and situational theories presented. Situational explanations will also be applied to each gender's attitude toward capitalism. A backlash hypothesis will be constructed to explain some of the difference of opinion over feminism and laissez faire economics. The final section of this chapter will conclude this study by summarising the various findings. Attention will be given to the implications that this research has for considerations of -69-gender differences in voting behaviour. Recommendations for a future research agenda focussing on situational differences between the genders will be given. Men and Reform: The Reform Party has more than its share of old soldiers and former cops (Flanagan, 1995). Until very recently, the military, police forces, fire department and a few other professions were one of the most abiding features of the gendered division of labour. Men were expected to fight invaders, criminals and blazes to protect women and children, who were barred from that fighting (Morgan, 1994). These perspectives were especially pronounced at early party conventions when the ideology of the party was developed, so it is unsurprising that a 'discourse of masculinity' has developed within the party on these issues. The capacity for violence is central to conceptions of masculinity projected from these roles, and the Reform party has adopted a hawkish stand on matters of defence and public security. A vengeance-based conception of justice, repealing gun control legislation, a referendum on capital punishment, increasing funding to defence spending, and support for NATO and NORAD are all part of the Reform party platform (Reform, 1997). The simplest explanation for men's support of forceful solutions to problems is that men are simply more accepting of violence than women due to socialised or even essential factors1. Social science literature has a long history of finding sex differences in aggression (Terman and Tyler, 1953; Whiting and Pope, 1973; Maccoby and Jacklin, 1974). Gilligan argues that men i Emile Durkhiem's dictum that social facts can be explained by social facts and not by biological causes still stands as valid. At present, there is no direct evidence for genetic control of specific human social behaviour (Gould, 1997:254), and the misuse of biology in 'new' fields like evolutionary psychology or anthropology is nothing more than revised social Darwinism. Millet writes that "patriarchy has a tenacious or powerful hold through its successful habit of passing itself off as nature" (Millet, 1990). -70-are more likely to see a violent sub-text to a wide variety of situations where women are not, especially within personal relationships (Gilligan 1982). Men are more aggressive in relations with other people, and aggression is perceived to be key to the masculine stereotype (Rosenkrantz, 1968). A recent work on Jungian psychoanalysis argues that "the Warrior is a basic building block of masculine psychology" (Moore and Gillette, 1990: 77). Sociologist Robert Brannon has identified several male stereotypes, including to be "venturesome, willing to take risks and engage in violence without hesitation" (1976). Socialised or even biological differences may play a role in the development of distinct attitudes in men and women in aggregate. But grand statements about the 'nature' of men and women are too blunt to dissect this phenomenon. A more profitable route for investigations into the causes of men's and women's voting behaviour may be in the rarely considered area of situational difference. In chapter 4, the need to investigate the role of 'private man' in politics was identified, in order to find out "what men fail to do altogether, what they claim to do symbolically rather than carry out practically" (Carver, 1996:4). Exploring the differences that actually exist in how men and women live their daily lives may prove to be a far richer field than the exhausted soil of essential or socialised difference. More insights into the gender gap are to be found in the changing of diapers than in DNA. Having primary responsibility for a child is as strongly related to 'feminine' traits, such as nurturance and sympathy, as being female. An experiment by sociologist Barbara Risman compared the personality traits of single fathers, single mothers, and married fathers, expecting to find that single and married fathers shared more traits due to biological differences or childhood socialisation. Instead, she found that single fathers and mothers shared more -71-'feminine' personality traits than single fathers and married fathers2 (Risman, 1987). In a similar study, men who provided care to ailing relatives were found to provide as much emotional support as women traditionally do. Men were found to engage in as much nurtuance -holding the sick relative's hand, listening, showing concern - as women in similar situations (Kaye and Applegate, 1990). The obligation for this kind of work normally falls upon women. But when men perform these important tasks, their attitudes and demeanor seems to become more caring. This may, in turn, produce more communitarian perspectives and greater opposition to violence. Gilligan hypothesises that women are more likely to adopt a code of ethics based on care and minimising distress, rather than abstract notions of justice (1983). Gidengil contends that nurturance and caring helps to account for women's opposition to the Canada-US Free Trade Agreement and its neo-liberal agenda (1995). In O'Neill's extensive study of the gender gap in Canada, men and women showed distinct belief systems, but her method of controls failed to account for the possibility of situational effects in men (1996). The gender gap on issues of compassion and violence appears to be linked to a caring-based ethic, but this may be the result of situational effects like responsibility for childcare. While these differences may be reinforced by socialised norms of behaviour, situational effects are a key intervening variable to test. Situational differences also help to explain the positioning of many men and women on the tension between the ideology of the market and the demands of social problems. Structural 2 The male sample was carefully controlled to ensure it did not draw an atypical group of nurturing men. Those selected had become single fathers in circumstances beyond their control, such as desertion or death of the mother. -72-approaches to this problem have been mostly unsuccessful and situational studies may help to explain the preponderance of men among Reform supporters. The Reform Party clearly subscribes to neo-conservative capitalist ideology. In 1995, Reform proposed a complete decentralization of social policy to the provinces, the elimination of equalization payments to the provinces and for unemployment and pensions plans to be replaced with personal, tax-sheltered savings accounts and private insurance (b:24; a:300; a:302-309). They blame government programs and the poor themselves for child poverty, saying that if "children are without food or clothing, it may be for reasons other than low family income... These (problems) are in part a direct result of harmful government policies in the first place" (a:298). The Reform Party advocates a return to laissez faire capitalism and. the abolition of the welfare state. Our question is why a disproportionate number of men would agree. An examination of the differences between 'private man' and 'private woman' illuminates the differing basis for poverty in each gender. Women are poor primarily due to the pressures and costs of caring for young children in the absense of adequate child care3. The same cannot be said of the majority of poor men. The pressures of raising children means that women work fewer hours in the paid labour force than men, and are disproportionately found in multiple, temporary or part-time jobs (CAC, 1994; Industry, 1995). The poverty rate for single mothers with children less than seven was 82.8 per cent in 1995, and these single mothers experienced deep poverty, with average incomes $8,851 below the poverty line (NCW, 1997). 3 Two indicators that women's employment and income are affected by the care of children are lower levels of employment among women with pre-school-age children compared to school-age children, and that single mothers have a substantially higher unemployment level than mothers in two parent families (Day, 1998; Industry, 1995). In 1996, 83 per cent of single parents were women (StatsCan). -73-However, the cost of parenting does not affect single mothers alone. Single fathers make significantly less than fathers in two-parent families, presumably due to the demands of child rearing (StatsCan, 1998). Several recent studies have shown that the time women spend out of the workforce to have children compromises their ability to remain competitive (O'Neill, 1994). "By the age of 64, women are away from their work an average of 14.7 years. Men are away an average of 1.6 years" (Van Weiss, 1994:92). These absences cost women seniority, experience, skills and training which are key in today's competitive marketplace. Other studies demonstrate the role cultural sexism has in undervaluing work in the 'caring' professions - teaching, nursing, child care or therapy. "Such jobs offer low pay relative to their requirements for education andskill and factors such as whether the job is unionized or involves physical hazards" (England and Folbre, 1997:1). Meanwhile, men are rewarded financially and socially for engaging in competitive behaviour, like professions in the financial or business community. Almost all women who have children will have lower lifetime earnings and less advancement than comparable men. Horatio Alger's tales of the wonders of capitalism seem more valid to a person who is not held back by responsibilities at home. Men's lack of action in their private lives leaves them free to compete in the public sphere. But women are still restricted by socially imposed roles and demands from families. Women's dubious attitudes toward capitalism are understandable if women can expect to earn less than men with the same education and ability because the woman will have to take time of work to reproduce and the man will not. Men's support for a competitive marketplace is understandable if they are relieved of obligations to act as parents -74-by social convention. Having the time to devote themselves to a career allows men to make better headway in society, realise goals more quickly and encounter fewer obstacles to success. Capitalism seems to work much better for men than women, so the relative difference in their attitudes is understandable. Rather than a 'masculinization of wealth' driving men to increasingly right-wing positions, men are enjoying 'masculinized opportunity.' An additional factor contributing to the male dominance of Reform Party support and the attitudinal differences detected in capitalism and feminism is Reform's populist backlash. Backlash can be defined as "the reaction by groups which are declining in a felt sense of importance, influence and power, as a result of secular endemic change in the society, to seek to reverse or stem the direction of change through political means" (Lipset, 1970:29). The Reform Party's brand of backlash is akin to the nativism of southern extremists in the civil rights-era, although directed against a variety of groups: feminists, environmentalists, visible minorities, Quebec nationalists and First Nations. Nativist sentiment is a resentment felt by those who share the ethnic, religious, racial (or gender) characteristics of the dominant class, but not the economic or political power (Harrison, 1995:7). Reform's membership is highly skewed to married, middle-aged, middle-class men; characteristics shared with captains of industry and government (Flanagan, 1995). While oppressed gender, race or language groups can appeal to the federal government for economic and social reforms based on notions of equity, nativist "deprivation tends to be manifest as sheer loss of status and influence; it cannot be assuaged directly by government action, and indeed an increment of government power typically means, for these groups, a diminution of their traditional power" (Lipset, 1970:23-24). As the market remains supportive of ethnic and gender -75-privilege, and government intervention tends to be directed against them, nativists are inclined to be economic libertarians. The Reform stand against feminism is a nativist expression of resentment against women demanding state action to achieve economic equity. Reform's position would be attractive to men who are covetous of gains made by women over the past thirty years and feel that they have been at their expense. Economic indicators show that the annual real earnings of the poorest ten percent of men declined 27 per cent between 1981 and 1993, and the annual real earnings of the middle decile of men have declined 9 per cent during the same period. Women have gained overall during this period, with poor women' real incomes rising 22 per cent and middle decile women's incomes rising by 11 per cent (Beaudry and Green, 1996). Under such conditions, programs specifically designed to alleviate women's poverty and systemic gender inequality in hiring or to promote equity will not be popular with men whose incomes are in decline. By attacking feminism, Reform attracts men upset about the changes they see around them and who want a return to the days when the 'family-wage' gave men enough economic clout to run a household. Part of the problem may be a conception of the world as a zero-sum game, where one person's gains are another's losses. Women's entrance into the workplace and higher education is new competition for men who traditionally occupied an elite place in society. With rising levels of unemployment over the past three decades and a decline in individual wages, arguments which blame women for high unemployment are growing more common among neo-conservatives: -76-'"Part of the unemployment is not as much recession,' Ronald Reagan said in a 1982 address on the economy, "as it is the great increase of the people going in the job market, and - ladies, I'm not picking on anyone but - because of the increase in women who are working today" (Faludi, 1991:67) This pattern of blaming government activism and the entrance of new workers for economic malaise is a phenomenon common to post-industrial advanced democracies. Kitschelt predicts that the highest incidence of support for new right parties like Reform will be among "craftspeople, shopkeepers, and blue-collar workers in industries that are losing ground in the international market competition." In Europe, these groups blame their economic problems on high taxes and immigration, and revert to authoritarian patriarchy oh issues of gender (1995:19). Reform appears to be a similar reactionary movement in Canada, and this economic backlash helps to explain part of the difference of opinion between the genders over capitalism. Conclusions: The gender gap is not a synonym for a women's voting block. Rather, gender differences in voting behaviour should be analysed in light of theories of coalitions and the relative influence of each sex over the question of government formation. The concept of a 'gender gap' - developed in the two-party US system - does not translate into the multi-party Canadian electoral regime nor does it accurately describe the relative power of each coalition. The size of the gender gap is irrelevant to electoral victory. The distance between the two genders is interesting only in terms of the sheer difference between men and women. Studies of the relative power of each gender must instead focus on the cohesion of each sex to a particular party, as the most cohesive gender is the one that will elect its preferred candidate. Power for each gender is a factor of its unity, not its disagreement with the other sex. The -77-elasticity of response to policy and patronage attention also advances the power of a voting block. Reconceptualising the gender gap in these terms allows consideration of the influence of men on election outcomes and a more balanced approach to the study of gender and elections. An exploration of the underlying differences in the political attitudes of men and women identified three ideological clusters that provide the basis of the gender gap in Reform Party support. Support for capitalism was found to be much higher among men than women, a fact noted in a variety of studies (Erie and Rein, 1988; Gidengil, 1995). However, this difference did not extend to other expected economic factors like support for the welfare state or continentalism. Of the variables tested, Feminism was found to be the only significant cultural difference between men and women. The largest single attitudinal cause of the gender gap was found to be issues surrounding the use of force. Men were much more supportive than women of government employing violent means. When combined, these three variables explain the entire gender gap in support for the Reform Party. Women have been slightly more cohesive than men during the period of 1974 to 1997 in Canada. However, women's greater influence over election outcomes was restricted primarily to the period of Pierre Trudeau's leadership of the Liberal Party of Canada. During the three elections tested which involved Trudeau, the average Liberal vote among women was 46.6 per cent, compared to 36.3 in the four elections after his departure. The Liberal majority governments formed under Jean Chretien owe less to the cohesion of women than those of Trudeau did. Women may have been attracted to the social justice policies and unconventional posture of Trudeau, and the business liberalism of recent Liberal leaders has squandered the -78-support of a disproportionate number of women. Exploring the effect of Pierre Trudeau on the gender gap may yield new empirical evidence of the power of gender appeal. Gender difference in Canadian's voting behaviour has shifted from the parties that contest to form governments to more extreme parties who influence from outside the governing coalition, like Reform and the New Democrats. In 1997, the NDP registered a significantly higher vote among women than men for the first time. While the Liberals are losing support among women and becoming gender neutral in their support, the conservative party of families is becoming increasingly male dominated. As some women leave the Liberal Party and their support becomes gender neutral, the stable level of cohesion that men express for conservative parties is becoming increasingly important. The electoral rift between the Progressive Conservative Party and the Reform Party prevents the male dominated group of conservative Canadians from forming the government. So long as these large numbers of men (and a significantly smaller group of women) continue to vote for a party too extreme to be elected nationally, conservative voters will be stymied. At present, women are also showing signs of fragmenting their support and shifting to the less moderate NDP, who also hold no chance of forming a federal government. If this trend continues, neither gender will be cohesive enough to elect a majority government. In order to exert maximum influence over policy, the gender block must shift to the centrist parties that can form governments, rather than counting on minority opposition parties to carry their issues in Ottawa. These centrist parties would also be wise to adopt policies that will attract back the predominant support of a gender. Genders are such large voting blocks that cohesive support from one is a virtually a ticket to form a majority government. -79-Conservative parties may choose to consciously appeal to men. In order to do this, issues of relevance to men will have to be identified. Conservative appeals are derived from a 'discourse of masculinity' and naturally appeal to more men then women. The challenge for progressive parties will be to mute this natural appeal by determining what issues in their platform have salience for men and stressing these issues to male audiences. Targeting potential male supporters is key for progressive parties in jurisdictions with salient gender cleavages, where conservative parties will target vulnerable women voters. This study points to crime, integrity, unity, and finances as potential 'men's issues.' But to understand what motivates men we have to look into their private lives. 'Private man' remains a mysterious figure in political science literature. While the 'public' political man has long been the focus of academic inquiry, more personal investigations into the home and family interactions of male actors have not been plentiful. Understanding the actions, demands and constraints placed on men helps to explain behaviour in more testable ways that situational or essential difference and with potentially greater rewards than the seemingly limited structural explorations. The examples listed here have included only child care and care for relatives, but can be extended to a wide-range of areas including the effect of women entering 'male' professions on male attitudes, the influence of media images on male self-esteem and the growing objectification of male bodies. However, there are presently few surveys that contain both questions of political attitudes and situational constraints. This oversight will have to be addressed if the relationship between situation and political behaviour is to be explored. -80-Bibliography Abramowitz, Alan I. and Kyle L. Saunders (1998). "Ideological Realignment in the U.S. Electorate." Journal of Politics 60:634-52. Abzug, Bella with Mini Kebler (1984). The Gender Gap. Boston, MA: Houghton Mifflin. Adams, Michael (1997). Sex in the Snow: Canadian Social Values at the End of the Millenium. Toronto: Penguin. Almond, Gabriel and Sidney Verba (1963). The Civic Cultures: Political Attitudes and Democracy in Five Nations. Princeton, NJ: Princeton University Press. Andersen, Kristi and Elizabeth Cook (1985). "Women, Work and Political Attitudes." American Journal of Political Science 29:606-625. Arrow, Kenneth J. (1951). Social Choice and Individual Values. New York: John Wiley. Axelrod, Robert (1972). "Where the Votes Come From: An Analysis of Electoral Coalitions, 1952-1968." American Political Science Review 66:12-20. Bacchi, Carol (1986). "Women and Peace Through the Polls." Politics: Journal of the Australasian Political Studies Association 21:62-67. Bacchi, Carol Lee (1990). Same Difference: Feminism and Sexual Difference. North Sydney, Australia: Allen & Unwin. Bakvis, Herman and Laura G. MacPherson (1995). "Quebec Block Voting and the Canadian Electoral System." Canadian Journal of Political Science 28:659-692. Barlow, Maude & Bruce Campbell (1995). Straight Through the Heart: How the Liberals Abandoned the Just Society. Toronto: HarperCollins. Bartolini, Stefano and Peter Mair (1990). Identity, Competition, and Electoral Availability: The Stabilisation of European Electorates 1885-1985. Cambridge: Cambridge University Press. Bashevkin, Sylvia B (1998). Women on the Defensive: Living Through Conservative Times. Toronto: University of Toronto Press. Beach, Charles M. and George A. Slotsve (1996). Are We Becoming Two Societies?: Income Polarization and the Middle Class in Canada. Toronto: CD Howe Institute. -81-Beaudry, Paul and David Green (1996). "Vision Paper: Individual Responses." Unpublished report for Industry Canada. Department of Economics, University of British Columbia. Berke, Richard L. (1998). "Democrats' Gains Dispel Notion That the G.O.P. Benefits From Low Turnout." New York Times. Thursday, November 4, 1998. p. 9. Bernard, Jessie (1981). The Female World. New York: The Free Press. Blake, Donald E. and Lynda Erickson (1998). "Electoral Volatility and Realignment in 1993." Delivered to the Annual Meeting of the Canadian Political Science Association, University of Ottawa, 1998. Bonk, Kathy (1980). "The Selling of the 'Gender Gap': The Role of Organized Feminism." In The Politics of the Gender Gap, edited by Carol Mueller. Beverley Hills, CA: Sage. Bourque, Susan C. and J. Grossholtz (1974). "Politics as Unnatural Practice: Political Science Looks at Female Participation." Politics and Society 4:225-66. Brannon, Robert (1976). "The male sex-role: Our culture's blueprint for manhood and what it's done for us lately." In The Forty-Nine Percent Majority: The Male Sex Role, edited by Deborah David and Robert Brannon. Reading, MA: Addison-Wesley. Brod, Harry (1987). "The Case for Men's Studies." In Changing Men: New Directions in Research on Men and Masculinity, edited by Michael Kimmell. London: Sage Focus. Burvis, Val (1984). "The Meaning of the Gender Gap: A Comment on Goertzel." Journal of Political and Military Sociology 12:335-343. Cairns, Alan C. (1968). "The Electoral System and the Party System in Canada, 1921-1965." Canadian Journal of Political Science 1:64-85. Campbell, Douglas, Philip E. Converse, Warren E. Miller, and Donald Stokes (1964). The American Voter: An Abridgement. New York: John Wiley and Sons. Campbell, Colin and William Christian (1996). Parties. Leaders, and Ideologies in Canada. Toronto:McGraw-Hill Ryerson. Canadian Advisory Council on the Status of Women (1994). Work in Progress: Tracking Women's Equality in Canada. Ottawa: Canadian Advisory Council on the Status of Women. Carabillo, Judith Mueli and June Bundy Csida (1993). Feminist Chronicles. 1953-1993. Los Angeles:Women's Graphics. -82-Carroll, S.J. (1988). "Women's autonomy and the gender gap: 1980 and 1982." In The Politics of the Gender Gap, edited by Carol Mueller. Beverley Hills, CA: Sage. Carver, Terrell (1996a). Gender is not a synonym for women. London: Lynne Rienner. Carver, Terrell (1996b). '"Public Man' and the Critque of Masculinities." Political Theory 24:673-686. Chambers, Veronica (1995). "Betrayal Politics." In Listen Up: Voices from the Next Feminist Generation, edited by Barbara Findlen. Seattle, WA: Seal Press. Chodorow, Nancy (1978). The Reproduction of Mothering. Berkeley, CA: University of California Press. Chretien, Jean (1986). Straight from the Heart. Toronto: Key Porter. Cicone, M.V. & Ruble, D.N. (1978). "Beliefs about Males." Journal of Social Issues 34:5-16. Clark-Stewart, K. (1980) "A Father's Contribution to children's cognitive and social development." In The Father-Infant Relationship, edited by F. Pedersen. New York: Praeger. Clymer, Adam (1980). "Displeasure with Carter turned many to Reagan." New York Times November 9. Connell, R.W. (1987). Gender & Politics. Stanford, California: Stanford University Press. Conover, P.J. (1988). "Feminists and the Gender Gap." Journal of Politics 50:985-1010. Cook, E.A. & Wilcox, C. (1991). "Feminism and the gender gap - a second look." Journal of Politics 53:1111-1122. Cornwall, Andrea and Nancy Lindisfarne (1994). Dislocating masculinity: Comparative ethnographies. London: Routledge. Courtney, John and David E. Smith (1966). "Voting in a Provincial General Election and a Federal By-election: A Constituency Study of Saskatoon City." Canadian Journal of Economics and Political Science. 32:338-53. Day, Shelagh and Gwen Brodsky (1998). Women and the Equality Deficit: The Impact of Restructuring Canada's Social Programs. Ottawa: Status of Women Canada. -83-Deitch, Cynthia (1988). "Sex Differences in Support for Government Spending." In The Politics of the Gender Gap, edited by Carol Mueller. Beverley Hills, CA: Sage. Dobbin, Murray (1991). Preston Manning and the Reform Party. Toronto: James Lorimer and Company. Downs, Anthony (1957). An Economic Theory of Democracy. New York: HarperCollins. Duverger, Maurice (1955). The Political Role of Women. New York: UNESCO. Ehrenreich, Barbara. (1983). The Hearts of Men. Garden City, New York: Anchor Press. Elshtain, Jean Bethke (1981). Public Man. Private Women. Oxford: Martin Robertson. Elshtain, J. B. (1984). "The Politics of Gender: Why some sound a different note." Progressive 4:28-49. England, Paula and Nancy Folbre (1998). "The Cost of Caring." Paper prepared for special issue of The Annals of the American Academy of Political and Social Science on Emotional Labor, edited by Ronnie J. Steinberg and Deborah M . Figart. February 18, 1998. Enloe, Cynthia (1990). Bananas, Beaches and Bases: Making Feminist Sense of International Politics. Berkeley: University of California Press. Erie, S.P.& Rein, M . (1988). "Women and the Welfare State." In The Politics of the Gender Gap, edited by Carol Mueller. Beverley Hills, CA: Sage. Evans, Judith (1986). Feminism and Political Theory. London: Sage. Everitt, Joanna (1994). "The Gender Gap on Social Welfare Issues: 1966-1990" Paper presented to the A G M of the Canadian Political Science Association. Calgary, Alberta. Everitt, Joanne (1997). "The Gender Gap in Canada: Now you see it, Now you don't." Unpublished manuscript. Faludi, Susan (1991). Backlash: The Undeclared War Against American Women. New York: Doubleday. Feldman, S. (1988). "The structure and consistency in public opinion: The role of core beliefs and values." American Journal of Political Science 32:416-440. -84-Fine, Terri Susan. (1993). "Social Position and Commitment to Core Values: Grappling with the (Non) Impact of Gender." Women & Politics 13:53-72. Flanagan, Tom. (1995). Waiting for the Wave: The Reform Party and Preston Manning. Toronto: Stoddart. \ Fox, Richard Logan. (1997). Gender Dynamics in Congressional Elections. Thousand Oaks: SAGE Publications. Francovic, Kathleen A. (1982). "Sex and Politics-New Alignments, Old Issues." Political Studies 15:439-48. Fraser, Graham (1996). "U.S. election reveals his and her politics". The Globe and Mail. November 8. p. A1-A14. Fukayama, Francis (1998). "Women and the Evolution of World Politics." Foreign Affairs 77:5. Gidengel, Elisabeth (1995). "Economic Man - Social Woman? The Case of the gender gap in support for the Canada-US Free Trade Agreement." Comparative Political Studies 28:383-408. Gidengil, Elisabeth & Andre Blais, Richard Nadeau and Neil Nevitte. (1998). "How we voted: the gender gap." The Globe and Mail. March 2, 1998. p. A17. Gilbert, L. (1985). Men in Dual Career Families: Current Realities and Future Prospects. Hillsdale, NJ:Lawrence Erlbaum. Gilligan, Carol. (1982). In a Different Voice. Cambridge, MA: Harvard University Press. Goertzel, Ted (1983). "The Gender Gap: Sex, Family Income and Political Opinions in the Early 1980's." Journal of Political and Military Psychology 11:209-222. Gould, Meredith and Rochelle Kern-Daniels (1977). "Towards a Sociological Theory of Gender and Sex." The American Sociologist 12:182-89. Gould, Stephen Jay (1977). "Biological Potentiality vs Biological Determinism." In Ever Since Darwin. New York: W.W. Norton & Co. Gould, Stephen Jay (1981). The Mismeasure of Man. New York: W.W. Norton & Co. Graham, Ron (1995). Al l the King's Horses. Toronto: MacFarlane, Walter & Ross. -85-Greenco, C. and Maccoby, E. (1986). "How different is the 'different voice'?" Signs 4:310-316. Hansard (1994). p. 8719. December 6, 1994. Paul Forseth, MP. Harrison, Trevor (1995). Of Passionate Intensity: Right Wing Populism and the Reform Party. Toronto: University of Toronto Press. Hart, Vicky (1989). "Gender and the Rise and Fall of Class Politics." New Left Review. 175. Hartely, R. (1959). "Sex-role pressures in the socialization of the male child." Psychological Reports. 5:457-468. Heyes, Cressida J. (1997). "Anti-Essentialism in Practice: Carol Gilligan and Feminist Philosophy" Hvapatia. 12:142-163. hooks, bell (1981). Ain't I a Woman?: Black Women and Feminism. Boston, MA: South End Press. hooks, bell. (1984). Feminist Theory: From Margin to Centre. Boston, MA: South End Press. Humm, Maggie (1992). Modern Feminisms: Political. Literary. Cultural. New York: Columbia University Press. Industry Canada (1995). Women in Canada: A Statistical Report. Ottawa: Industry. Inglehart, Ronald (1990). Culture Shift in Advanced Industrial Society. Princeton, NJ: Princeton University Press. Jackson, David (1990). Unmasking Masculinity: A Critical Autobiography. London: Unwin Hyman Ltd. Jamieson, Kathleen Hall (1995). Beyond the Double Bind: Women and Leadership. New York: Oxford University Press. Janeway, Elisabeth (1975). Between Myth and Morning: Women Awakening. New York: > William Morrow. Jenkinson, Michael. (1996). "He's Not as Cuddly as Clinton". Alberta Report November 25, 1996. p.12-14. -86-Johnston, Richard, Andre Blais, Henry E. Brady, and Jean Crete (1991). Letting the People Decide: Dynamics of a Canadian Election. Montreal: McGill-Queen's University Press. Johnston, Richard, Andre Blais, Henry E. Brady, Elisabeth Gidengil and Neil Nevitte (1996). "The 1993 Canadian General Election: Realignment, Dealignment or Something Else?" Paper presented at the Canadian Political Science Association meeting, Brock University. Johnston, Richard, Andre Blais, Elisabeth Gidengil and Neil Nevitte (1996a). The Challenge of Direct Democracy: The 1992 Canadian Referendum. Montreal: McGill-Queen's University Press. Jones, L.C. (1985). "Father-infant relationships in the first year of life." In Dimensions of Fatherhood, edited by S. Hanson & F. Bozett. Newbury Park, CA: Sage. Jump, Teresa L. & Linda Haas (1987). "Fathers in Transition: Dual-Career Fathers Participating in Child Care." In Changing Men: New Directions in Research on Men & Masculinity, edited by Michael Kimmell. Newbury Park, California: Sage Publishing. Kaye, Lenard W. and Jeffery S. Applegate (1990). "Men as elder caregivers: A response to changing families." American Journal of Orthopsychiatry 60:86-95. Kenski, Henry (1988). "The Gender Gactor in a Changing Electorate." In The Politics of the Gender Gap, edited by Carol Mueller. Beverley Hills, CA: Sage. Kilian, Crawford (1998). "Why are Boys Dumber?" Geogia Straight. February 12-19, 1998. Kimmel, Michael S. (1987). "Rethinking 'Masculinity': New Directions in Research." In Changing Men: New Directions in Research on Men & Masculinity, edited by Michael Kimmell. Newbury Park, California: Sage Publishing. Kirchten, D. (1984). "The Reagan reelection campaign hopes 1984 will be the year of the women." National Journal. 16:1082-1085. Kitschelt, Herbert (1995). The Radical Right in Western Europe: A Comparative Analysis. Ann Arbor, MI: University of Michigan Press. Klien, Ethel (1985).Gender Politics. Cambridge, MA: Harvard University Press. Kopinak, Katherine (1987). "Gender differences in political ideology in Canada." Canadian Review of Sociology and Anthropology 24:23-38. -87-LaPonce, Jean (1969). People versus Politics. Toronto, ON: University of Toronto Press. Liberal Party of Canada (1997). Securing Our Future Together: Preparing Canada for the 21st Century. (Red Book II). Ottawa: Federal Liberal Agency of Canada. Lipset, Seymour Martin and Earl Raab (1970). The Politics of Unreason. New York: Harper and Row. Littleton, Christine (1981). "Toward a Redefinition of Sexual Equality". Harvard Law Review. 95. pp. 487-508. Lovenduski, Joni (1981). "Toward the Emasculation of Political Science: The Impact of Feminism" in Men's Studies Modified, edited by Dale Spender. Willowdale, Ontario: Pergamon. Maccoby, Eleanor and Carol Jacklin (1974). The Psychology of Sex Differences. Stanford, CA: Stanford University Press. Mackie, Marlene (1991). Gender Relations in Canada: Further Explorations. Toronto: Butterworths. Mackie, Richard (1996). "Women spurn Ontario Tories." The Globe and Mail. December 30, 1997. A1,A4, MacBride-King, Judith L. (1990). Work and Family: Employment Challenges of the '90s. Ottawa: Conference Board of Canada. McCormack, Thelma (1975). "Towards aNonsexist Perspective on Social Change" in Another Voice: Feminist Perspectives on Social Life and Social Science, edited by Marci Millman and Rosabeth Moss. Garden City, NY: Anchor Books. McDanial, Susan (1988). "The Changing Canadian Family: Women's Roles and the Impact of Feminism." In Changing Patterns: Women in Canada, edited by S. Burt, L. Code and L. Dorney. Toronto: McClelland and Stewart Inc. McMenemy, John (1976). "Parliamentary Parties." In Political Parties in Canada, edited by C. Winn and J. McMenemy. Toronto: McGraw-Hill Ryerson. Messner, Michael A. (1998). The Limits of "The Male Sex Role": An Analysis of the Men's Liberation and Men's Rights Movements' Discourse. Gender and Society 12:255-276. Miller, Arthur (1988). "Gender and the vote: 1984." In The Politics of the Gender Gap, edited by Carol Mueller. Beverley Hills, CA: Sage. -88-Mishkind Marc E., Judith Rodin, Lisa R. Silberstien, and Ruth H. Striegal-Moore (1987). "The Embodiment of Masculinity: Cultural, Psychological and Behavioural Dimensions." In Changing Men: New Directions in Research on Men & Masculinity, edited by Michael Kimmell. Newbury Park, California: Sage Publishing. Moore, Robert and Douglas Gillette (1990). King, Warrior, Magician, Lover: Rediscovering the Archetypes of the Mature Masculine. New York: HarperCollins. Morgan, David H. J. (1994). "Theatres of War: Combat, the Military and Masculinities." In Theorizing Masculinities, edited by Harry Brod and Michael Kaufman. Thousand Oaks, CA: SAGE Publications. National Council of Welfare (1997). Poverty Profile. 1995: A Report by the National Council of Welfare Ottawa: Supply and Service Canada. New Democratic Party of Canada (1997). Framework for Canada's Future. New Democrats of Canada Association. Norris, Pippa (1986). "The Gender Gap: A Cross-National Trend?" In The Politics of the Gender Gap, edited by Carol Mueller. Beverley Hills, CA: Sage. Northrup, David A. (1998). The 1997 Canadian Election Study: Technical Documentation. Institute for Social Research, York University. Okin, Susan Moller (1979). Women in Western Political Thought. Princeton, NJ: Princeton University Press. O'Neil, James M. , Glenn E. Good, and Sarah Holmes (1995). "Fifteen Years of Theory and Research on Men's Gender Role Conflict: New Paradigms for Empirical Research." In A New Psychology of Men, edited by Ronald F. Levant and William S. Pollack. New York: HarperCollins. O'Neill, Brenda(1996). Rethinking Political Thinking: Gender and Public Opinion in Canada. Ph.D. Dissertation. Department of Political Science, University of British Columbia. O'Neill, Tip and William Novak (1987). Man of the House. New York: Random House. Orr, Catherine M . (1997). "Looking for the Third Wave." Hyapatia 3:30-44. Pateman, Carole and Teresa Brennan (1979). "Mere Auxiliaries to the Commonwealth". Political Studies 27:183-200. Pateman, Carole (1988). The Sexual Contract. Cambridge: Polity Press. -89-Petersen, V. Spike, ed (1992). Gendered States: Feminist (Re)visions of International Relations Theory. Boulder: Westview. Peterson, V. Spike and Jacqui True (1998). '"New Times' and New Conversations." In The "Man" Question in International Relations, edited by Marysia Zalewski and Jane Parpart. New York: HarperCollins. Phillips, Anne (1992). "Universal Pretensions in Political Thought." In Destablising Theory, edited by Michele Barrett and Anne Phillips. Cambridge: Polity Press. Philp, Margaret (1998). "Women gaining in wage-equality contest." Globe and Mail March 24. Pleck, Joseph H. (1987). "American Fathering in Historical Perspectives." in Changing Men: New Directions in Research on Men & Masculinity, edited by Michael Kimmell. Newbury Park, California: Sage Publishing. Pomper Gerald M . (1977). Voter's Choice. New York: Dodd, Mead. Poole, Keith T. and Harmon L. Zeigler (1985). Women. Public Opinion and Politics. New York: Longman. Progressive Conservative Party of Canada (1997). Let the Future Begin: Jean Charest's Plan for Canada's Next Century. Ottawa: PC Canada Fund. Rae, Douglas W. and Michael Taylor (1970). The Analysis of Political Cleavages. New Haven, CT: Yale University Press. Rebick, Judy (1998). "The Silent Treatment." Elm Street. April, p. 102. Reform Party of Canada (1995a). Reform Party Dissenting Opinion for the Standing Committee on Human Resources Development. Ottawa: Public Works Canada. Reform Party of Canada (1995b). Taxpayer's Budget. Ottawa: Public Works Canada. Reform Party of Canada (1997). Blue Book: 1996-1997 Principles & Policies of The Reform Party of Canada. Calgary, AB: Reform Fund Canada. Reform Party of Canada (1997b). Reform Youth. Calgary, AB: Reform Fund of Canada. Reform Watch (1996). "Reform vs Women." Reform Watch 18. October. -90-Rhodebeck, Laurie A. (1996). "The Structure of Men's and Women's Feminist Orientations: Feminist Identity and Feminist Opinion" Gender and Society 10:386-403. Rinehart, Sue Tolleson (1992). Gender Consciousness and Politics. New York: Routledge. Riker, William H. (1962). The Theory of Political Coalitions. New Haven, CT: Yale University Press. Risman, Barbara J. (1987). "Intimate Relationships from a micro structural perspective: Men who mother". Gender and Society 1:6-32. Rosenkrantz, P., S. Vogel, H. Bee, I. Broverman & D. Broverman (1968). "Sex-role stereotypes and self-concepts in college students." Journal of Sonsulting and Clinical Psychology 32:287-295. Sapiro, Virginia (1990). "The Women's Movement and the Creation of Gender Consciousness: Social Movements as Socialisation Agents." In Political Socialization. Citizenship. Education and Democracy, edited by Orit Ichilov. New York: Teachers College Press. Schafly, Phyllis (1996). "What Caused the Gender Gap?" Eagle Forum. December 12. p. 2. Schultz, Vicki (1998)."Reconceptualising Sexual Harrassment" Yale Law Journal. Vol. 107, No. 6. Schwartz, Amy E. (1996). "Gender gap creates fallacies about the female voter." The Washington Post. November 24, 1996. Seltzer, Richard A., Jody Newman and Melissa Voorhees Leighton 1997. Sex as a Political Variable: Women as Candidates and Voters in U.S. Elections. Boulder, Colorado: Lynne Rienner Publishers, Inc. Shapiro, Robert and Harpreet Mahajan (1986). "Gender Differences in Policy Preferences: A Summary of Trends from the 1960s to the 1980s." Public Opinion Quarterly. 50:42-61. Shapley,L.S.(1953). "A Value for N-Person Games." Annals of Mathematics Study. 28:308-317. Sharpe, Sydney and Don Braid (1992). Storming Babylon: Preston Manning and the Rise of the Reform Party. Toronto: Key Porter Books. Sharma, Nandita (1997). The National Action Committee on the Status of Women's Voters' Guide: A Women's Agenda for Social Justice. Toronto: James Lorimer and Company. -91-Sinclair, Peter R. (1979). "Class structure and populist protest: The case of Western Canada." In Society and Politics in Alberta: Research Papers, edited by C. Coldarola. Toronto: Methuen. Smeal, Eleanor (1984). How and Why Women Will Elect the Next President. New York: Harper &* Row. Stimpson, Catharine R. (1984). "Woman as Knowers." In Feminist Visions, edited by D.L. Fowlkes and C.S. McClure. University, Alabama: University of Alabama Press. Strauss, Steven (1998). "The Boys of War." Globe and Mail. October 10, 1998. D5. Taylor, Jill, Carol Gilligan and Amy M. Sullivan (1995). Between Voice and Silence: Women and girls, race and relationship. Cambridge, MA: Harvard University Press. Tavris, Carol (1992). The Mismeasure of Woman. New York, NY: Simon and Shuster. Terman, L. and Tyler L. (1954). "Psychological Sex Differences." In Manual of Child . Psychology. Second Edition, edited by L. Carmichael. New York: John Wiley. Tickner, J. Ann (1992). Gender in International Relations: Feminist Perspectives on Achieving Global Security. New York: Columbia University Press. Thompson, Edward H., Jr. and Joseph H. Pleck (1987). "The Structure of Male Role Norms." In Changing Men: New Directions in Research on Men & Masculinity, edited by Michael Kimmell. Newbury Park, California: Sage Publishing. Togeby, Lise (1994). "Politial implications of increasing numbers of women in the labor force." Comparative Political Studies 27:211-240. Valpy, Micheal (1998). "Guardians of good government." Globe and Mail. March 28. D9. Van Loon, Rick (1970). "Political Participation in Canada." In Canadian Political Behaviour: Introductory Readings, edited by Sylvia B. Bashevkin. Toronto:Methuen. Walker, Lawrence J. (1997). "Is Morality Gendered in Early Parent-Child Relationships? A Commentary on the Lossis, Ross, and Leroux Study." Merrill-Palmer Quarterly. 43:148-159. Weiler, Marcus (1987). "Neo-conservatism: A Study of Ideological Argument." Journal of the American Forensic Association 24:37-47. -92-Wearing, Peter & Wearing, Joseph (1991). "Does gender make a difference in voting behaviour?" In The Ballot and its Message, edited by Joe Wearing. Toronto:Copp, Clark, Pittman. Whiting, Beatrice and Carolyn Pope (1973). "A Cross-cultural Analysis of Sex Differences in the Behaviour of Children Age Three to Eleven." Journal of Social Psychology 91:171-188. Wiseman, Nelson (1986). "The Use, Misuse, and Abuse of the National Election Studies." Journal of Canadian Studies 21:21-35. Zalewski, Marysia and Jane Parpart (1998). The "Man" Question in International Relations. New York:HarperCollins. -93-Appendix One: The Refinement of Data from the 1974-1988 CNES Studies The Canadian National Election Study provides a complex series of interwoven questions that are suitable for our study. Unfortunately, the numbers from all CNES research before 1988 suffer from problems in the data collection process that have produced results that are "unreliable, misleading and of questionable use" (Wiseman, 1986:25). When employing in-home clustered survey administration techniques, the CNES study was in the field as long as six months after voting day. Post-election events would then sway voters to indicate false support, especially those who had not actually cast ballots. In the extreme example, the 1979 study was not even completed before the short-lived Clark government fell. These methodological problems produced results that were obviously different from aggregate results collected by Elections Canada. According to the 1974 study, the Liberals received 52.4 per cent of the vote, when the aggregate results give them only 43 per cent. However, inaccurate vote reporting is unlikely to be affected by gender. Women and men did not differ significantly in turnout for these elections, and nonvoters contribute to the majority of the reporting error. The difference in support for each party between genders thus may be accepted as accurate. In order to refine the results in this study to more accurately reflect aggregate vote, for each party, the support for each gender will be lowered or raised by the discrepancy between the reported vote and the actual total votes as compiled by Elections Canada. This formula was employed: Survey Party Vote [Male or Female] - (Survey Party Vote - Actual Party Vote) = Actual Party Vote [Male or Female] Those parties that were under-reported in the CNES will be raised to reflect their actual support, while those who were over-reported will be reduced. As a result, the reported vote for each -94-party from men and women is now more accurate, although the gender gap between the two figures has remained undisturbed. While used on all elections between 1974-93, this refinement process was not used on the 1997 election for a number of reasons. The introduction of CATI (Computer-Assisted Telephone Interviewing) technology in the 1988 survey eliminated part of the over-reporting problem. By 1993, the margin between reported and actual results had become insignificant in most cases, lessening the need for this type of correction. The difference between reported and actual party vote was insignificant by 1997. More important, the various indices, demographic characteristics and opinion scales are developed using 1997 CNES data. The results for each party in 1997 will be left raw to ensure continuity across all the variables and indexes. This type of refinement was used on several tables. The gender gap for each table was left stable, but the level of support for parties varied widely in some cases. These changes altered the findings for the section on cohesion fundamentally. The process of refinement was determined before analysis of cohesion began, to prevent unconscious manipulation of the data. But without refinement, the CNES data was unusable. Trudeau did not win the 1974 election with 52.4 per cent of the vote, as the CNES would tell you, but with 43 per cent. Had the calculations of cohesion been made using the numbers provided, the levels of cohesion for the Liberal party would have artificially high compared to the Tories and NDP. By refining the CNES data, we are able to make more realistic appraisals of the state of the gender gap during the period in question. -95-Table 2.2 Canadian Federal Election Results 1974-1988 divided by Gender (Unrefined Data): Year NDP Men NDP Women NDP Gap Lib. Men Lib. Women Lib. Gap PC Men PC Women PC Gap 1974 13.4 9.6 -3.8 48 56.8 8.8 33.2 28.3 -4.9 1979 14.3 14 -0.3 38.2 46 7.8 43.9 34.6 -9.3 1980 15.4 17 1.6 43 52.1 9.1 40.6 28.8 -12 1984 15.3 15.3 0 23.3 26.7 3.4 58.9 55.8 -3.1 1988 16.7 17.7 1.0 22.4 29 6.6 46.2 38.6 -7.6 Data from Wearing, 1991. Table 5.4 Electoral Cohesiveness: Mean Percentage Vote for the Largest Party 1974-93 (Unrefined Data) Year Winning Party (Maj ority/Minority) Largest Vote Male (Party) Largest Vote Female (Party) Cohesion Male Cohesion Female. 1974 Liberal 48 (L) 56.8 (L) 10 18.8 1979 Conservative 43.9 (C) 46.0 (L)* 5.9 -3.4 1980 Liberal 43.0 (L) 52.1 (L) 5.0 14.1 1984 Conservative 58.9 (C) 55.8 (C) 20.9 17.8 1988 Conservative 46.2 (C) 38.6 (C) 8.2 0.6 1993 Liberal 40.3 (L) 44.1 (L) 2.3 6.1 1997 Liberal 37.1 (L) 37.4 (L) -0.9 -0.6 Mean 45.3 45.6 7.3 7.6 Taken from Canadian National Election Studies. Cohesion = distance from 38%. * Not Election Winner. -96-Appendix Two - The Construction of Attitudinal Indexes: All quantitative research in this study, with a noted exceptions, comes from the 1997 Canadian National Election Study. Data from the 1997 Canadian Election Survey were provided by the Institute for Social Research, York University. The survey was funded by the Social Sciences and Humanities Research Council of Canada (SSHRC), grant number 412-96-0007 and was completed for the 1997 Canadian Election Team of Andre Blais (Universite de Montreal), Elisabeth Gidengil (McGill University), Richard Nadeau (Universite de Montreal) and Neil Nevitte (University of Toronto). Neither the Institute for Social Research, the SSHRC, nor the Canadian Election Survey Team are responsible for the analyses and interpretations presented here. Variable Construction Questions Gender (Dichotomous Variable)-compute gender=9. if (mbsj2 eq l)gender=0. if (mbsj2 eq 2)gender=l. if (rtype3 ne l)gender=9. missing values gender (9). Capitalism Scale-compute capital=0. if (mbsa4 eq 3 or mbsa4 eq 4)capital=capital+l. if (mbsalO eq 1 or mbsalO eq 2)capital=capital+l. if (mbsal2 eq 1 or mbsal2 eq 2)capital=capital+l. if (mbsbl eq 2)capital=capital+l. if (mbsb5 eq l)capital=capital+l. if (mbsb6 eq l)capital=capital+l. if (mbsdlO eq 1 or mbsdlO eq 2)capital=capital+l. compute capital=capital/7. if (rtype3 ne l)capital=9. missing values capital (9). Cronbach's Alpha = .3076 Continentalism Scale-compute continen=0. if (mbsal5 eq 3 or mbsal5 eq 4)continen=continen+l. if(mbsdll eq 1 ormbsdll eq 2)continen=continen+l. if (mbsglO eq 1 or mbsglO eq 2)continen=continen+l. compute continen=continen/3. if(rtype3 ne l)continen=9. missing values continen (9). Cronbach's Alpha = .4566 Gov do more reduce income gap Rich&Poor Not Big Problem Some Have More Chance People Really Want Work, Can Find a Job Government Should <Standard of Living> Closer to Your View <People on Welfare> People in Government <Waste Tax Money> Unemployed Move to Regions Where Jobs Domore Protect CDN Business from Foreign International Trade Creates More Jobs Free Trade with US has been Good Anti-Welfare Scale-compute antiwelf=0. if (cpsela eq l)antiwelf=antiwelf+l. if(cpsj6eq l)antiwelf=antiwelf+l. if(pese6b eq l)antiwelf=antiwelf+l. if (pese6d eq l)antiwelf=antiwelf+l. if (pese6e eq l)antiwelf=antiwelf+l. compute antiwelf=antiwelf/5. if(rtype3 ne l)antiwelf=9. missing values antiwelf (9). Cronbach's Alpha = .4922 Cut Taxes=Cut Social Programs/Increase=Improve Federal Spending Cuts Been Fair/Unfair Cut Spending>Welfare Cut Spending>HealthCare Cut Spending>Unemployment Insurance -97-Macro-economics Scale-compute macro=0. if (cpsf4 eq 1 or cpsf4 eq 2)macro=macro+l. if (cpsf5 eq 1 or cpsf5 eq 2)macro=macro+l. if (cpsf6 eq 1 or cpsf6 eq 2)macro=macro+l. compute macro=macro/3. if (rtype3 ne l)macro=9. missing values macro (9). Cronbach's Alpha = .4735 Not much Any Gov do Solve Unemployment Maintain Social Programs=Eliminate Deficit Gov Leave to Private Sector Create Jobs Moral Traditionalism Scale-compute moral=0. if (mbsal eq 1 or mbsal eq 2)moral=moral+l. if (mbsa2 eq 3 or mbsa2 eq 4)moral=moral+l. if (mbsa7 eq 1 or mbsa7 eq 2)moral=moral+l. if (mbsa8 eq 3 or mbsa8 eq 4)moral=moral+l. if (mbsa9 eq 1 or mbsa9 eq 2)moral=moral+l. if (mbsg3 eq 3 or mbsg3 eq 4)moral=moral+l. compute moral=moral/6. if (rtype3 ne l)moral=9. missing values moral (9). Cronbach's Alpha = .3062 Gone Too Far Pushing Equal Rights Be More Tolerant People Choose Standards New Liefstyle Contrib Breakdown Society Change=Adapt Our View of Moral Behaviour Fewer Problems = Traditional Family Values Homosexual Couples Allowed Legally Marry Feminism Scale-compute feminism=0. if (mbsa3 eq 3 or mbsa3 eq 4)feminism=feminism+l. if (mbsa5 eq 1 ormbsa5 eq 2)feminism=feminism+l. if (mbsb4 eq l)feminism=feminism+l. if (mbse2 eq l)feminism=feminism+l. if (mbse4 eq l)feminism=feminism+l. if (mbse7 eq l)feminism=feminism+l. compute feminism=feminism/6. if(rtype3 ne l)feminism=9. missing values feminism (9). Cronbach's Alpha = .3457 Lay Off Women Whose Husbands Have Jobs Difficult Women Get Jobs Equal Abilities When It Comes to Hiring <Quotas> Your Opinion > Feminist Movement Your View > Equality of Men & Women Feminist Movement Encourages Women: Anti-Ethnic Scale-compute antiethn=0. if (mbsal 1 eq 1 or mbsal 1 eq 2)antiethn=antiethn+l. if (mbsal4 eq 3 or mbsal4 eq 4)antiethn=antiethn+l. if (mbsd6 eq 1 or mbsd6 eq 2)antiethn=antiethn+l. if (mbsdl2 eq 3 or mbsdl2 eq 4)antiethn=antiethn+l. if (mbsg4 eq 1 or mbsg4 eq 2)antiethn=antiethn+l. if (mbse8 eq l)antiethn=antiethn+l. compute antiethn=antiethn/6. if(rtype3 ne l)antiethn=9. missing values antiethn (9). Cronbach's Alpha = .4142 Look After Cdns Born here First, Others 2nd Minority Groups Need Special Rights Parties Spend Too Much Time re Minorities Immigrants make Important Contribution Too Many Recent Immigrants Not Want Fit -98-Anti-French Scale-compute antifren=0. if (mbsd7 eq 1 or mbsd7 eq 2)antifren=antifren+l. if (mbsg6 eq 3 or mbsg6 eq 4)antifren=antifren+l. if (mbsg7 eq 3 or mbsg7 eq 4)antifren=antifren+l. if (mbsg8 eq 3 or mbsg8 eq 4)antifren=antifren+l. compute antifren=antifren/4. if(rtype3 ne l)antifren=9. missing values antifren (9). Cronbach's Alpha = .3248 Gone Too Far Pushing Bilingualism Anglos in Que Better Treated Fr in ROC Quebec has Right to Separate Quebec Separates, Aboriginals Remain Part Populism Scale-compute populism=0. if (mbsdl eq 3 or mbsdl eq 4)populism=populism+l if (mbsd2 eq 1 or mbsd2 eq 2)populism=populism+l if (mbsd3 eq 1 or mbsd3 eq 2)populism=populism+l if (mbsd4 eq 3 or mbsd4 eq 4)populism=populism+l if (mbsd5 eq 1 or mbsd5 eq 2)populism=populism+l if (mbsh5 eq 1 or mbsh5 eq 2)populism=populism+l compute populism=populism/6. if(rtype3 ne l)populism=9. missing values populism (9). Cronbach's Alpha = .4403 Most People not know what Best for Them People Have sense Tell Gov't Do Good Job Solve National Prob=Grassroots Decisions Gov ShldPay Most Attention Well-informed All Federal Parties Basically the Same Parties Spend Too Much Time Re Minorities Alienation Scale-compute alienate=0. if (cpsb9 eq 5 or cpsb9 eq 7)alienate=alienate+l. if(cpsbl0aeq 1 orcpsblOa.eq 2)alienate=alienate+l. if(cpsblOb eq 1 orcpsblOb eq 2)alienate=alienate+l. if(cpsblOd eq 1 orcpsblOd eq 2)alienate=alienate+l. if (cpsblOe eq 1 or cpsblOe eq 2)alienate=alienate+l. if (mbse9 eq l)alienate=alienate+l. compute alienate=alienate/6. if (rtype3 ne l)alienate=9. missing values alienate (9). Cronbach's Alpha = .4178 Satisfaction>Way Democracy Works inCanada Elected to Parliament Lose Touch People Poeple Like Me Not Have Say What Gov Does Politics&Gov Seem So Complicated Not Think Gov't Cares What People Think Politicians Ready to Lie to Get Elected Use of Force Scale-compute useforce=0. if(mbsgl eq 1 ormbsgl eq 2)useforce=useforce+l. if (mbse6 eq 2)useforce=useforce+l. if (pesel3 eq 5 or pesel3 eq 7)useforce=useforce+l. if (pesel2 eq 5 or pesel2 eq 7)useforce=useforce+l. if (pese6a eq 5)useforce=useforce+l. compute useforce=useforce/5. if(rtype3 ne l)useforce=9. missing values useforce (9). • Cronbach's Alpha = .2343 Participate Peacekeeping Even If Risk Your View > Marital Violence Capital Punishment Never Justified Only Police and Military Allowed Have Guns Cut Spending > Defense -99-

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0089138/manifest

Comment

Related Items