Scarcity of time or interest? Structural and cultural constraints on women s political participation

Transcription

1 Scarcity of time or interest? Structural and cultural constraints on women s political participation Rossella Ghigi Dipartimento di Scienze dell Educazione, Università Alma Mater Bologna (rossella.ghigi@unibo.it) Laura Sartori Dipartimento di Scienze Politiche e Sociali, Università Alma Mater Bologna (l.sartori@unibo.it) Dario Tuorto Dipartimento di Scienze dell Educazione, Università Alma Mater Bologna (dario.tuorto@unibo.it) 1. Introduction It is cross-nationally acknowledged that women are still far from having reached men s level of involvement in political actions, influence and representation (Burns 27; Burns et al. 1997; Norris 22; Paxton et al. 27; Schlozman et al. 1999), nor they equal in political knowledge and information (Burns et al 21; Dow 29). Political participation is the first arena where gender gap is still persistent, with women less politically engaged than men, and could be further emphasized by other cleavages in the social stratification structure (such as class, education or age, v. Burns 27; Norris 22). A number of explanations have been considered to address the gender dimension of political participation, since it is the outcome of multiple gendered dynamics such as socialization, attitudes, resources and opportunities (Bernstein 25; Conway et al. 24; Clark and Clark 1986; Fridkin and Kenney 27; Stake 27; Burns 27; Fridkin and Kenney 27; Lovenduski 25; Rapoport 1981). Life-cycle transitions in partnership and parenthood as well can cause transformations in economic resources and time availability, which have been proved to affect political participation (Kingston and Finkel, 1987; Sayer 25; Amato, 2; Kalmijn and Broese van Groenou, 25; Stoker and Jennings, 1995). Other researches on gendered political participation take into account different types of engagement. More institutional forms of participation such as voting, working on a campaign, and joining a party organization diverge from more activist practices such as buy or boycott products or signing a petition in the material or symbolic resources they require from participants, and since men and women have a different access to these resources, they are led toward different forms of engagement (Lister 23; Lovenduski 1998; Burns 27; Lister 23; Paxton et al. 27) preferring forms that can be incorporated in daily life, such as buy/boycott products (Stolle et al. 25). Coffé and Bolzendahl (21) combine the two perspectives (demographic and attitudinal characteristics of participants with different types of engagement) by investigating gender gaps in political participation in 18 advanced Western democracies. They show that women are more likely 1

2 than men to vote and engage in private activism, while men seem to prefer direct contact-based participation, collective types of actions and are (more active) members of political parties. In sum, there are two dominant set of explanations as to why women have not the same political involvement as men: they are unable to participate (structural constraints), and they are not interested in it (cultural constraints). Taking into account this literature, we are going to test four main hypotheses. 1. Women do participate less than men in political activities. 2. If political participation occurs, women prefer non-conventional forms. 3. Women don t participate primarily due to scarcity of time, as family constraints play a major role in lowering participation s rates both in political and non-political activities. 4. Women don t participate primarily due to scarcity of interest. Cultural constraints play a major role in differentiating and gendering the socialization to politics. When this cultural conditioning is not at work, and women have high political interest, their participation in collective and private activism is not penalized by familiar constraints. The data sources used for the study are: a) Indagine Multiscopo Aspetti della vita quotidiana Istat (211); and b) Itanes post-election survey (213). The first dataset is not specifically related to electoral or political study but includes an adequate set of variables related to political behaviour based on a large sample. Itanes survey covers a relatively small sample (3 cases) but allows for a wide range of indicators on politics with basic information on family status and composition. 2. Dimensions of political participation Istat (211) dataset provides information on three different indicators of participation: conventional participation, occurring within the parties (such as simple inscription, volunteering, distribution of leaflets, donations); mobilization (internal or external to the parties, taking place in a collective sphere); and invisible participation (which refers to the propensity to be interested, inquire and discuss politics). Itanes (213) dataset, instead, focus on different forms of political action and attitudes that account for collective and private forms of political activism. We then also report data about non-political participation such as participation to trade-unions, volunteering activities, ecology, peace and civil rights associations, cultural and recreational associations and professional associations. In this paper we will thus refer to Istat dataset when investigating two dimensions of political participation: the visible political involvement (measured by conventional participation 2

3 plus different forms of mobilization) and the invisible political involvement. On the other side Itanes will help us understanding the dichotomy collective vs. private activism in order to interpret eventual gendered dynamics. Our findings reported in Table 1 show that men are more likely to participate to any form of political participation, from activities related to parties to different actions of political mobilization, especially for attending political meetings. It s noteworthy, though, that the overall involvement in this kind of activities regards a small percentage of the sample. Concerning invisible involvement, women result significantly less active than men, both in terms of political information and propensity to discuss and listen to political debates. The gender difference is almost 1 percentage points higher for men when listening to political debates (27% of men and 19% of women), and 12 percentage point for women when it comes to information (16% of men and 28% of women never get informed) and to political discussion (22% and 4%). Tab. 1 Percentage of people that engaged in political activities at least once in the last 12 months, by gender see appendix Tab. 2. Percentages of men and women who participated in political parties activities, political mobilization, and got informed on politics by principal socio-demographic characteristics In a more detailed level of analysis, Table 2 highlights a decreasing gap from older cohorts to younger ones, both for visible and invisible forms of involvement. The gender gap is significantly reduced as the level of education increases, thus supporting a stronger relationship between participation and education for women. In other words, the difference in political involvement between men and women is higher at lower educational levels, but it then vanishes among the highly educated. With regard of occupational status, it can be noted that women out of job market (above all, those retired) are the most penalized in terms of visible as well as invisible political involvement. Self-employed people and working class show higher levels of gender gap. The gap is also high in Centre and Southern Italy, while in other areas of the country the gap is always present (even where local subcultures are strong) but in a much more nuanced form. Itanes (213) dataset focuses on political actions (neither affiliation nor usual political behaviours, as for Istat) and confirms a gendered set of practises and attitudes. Men are more likely to be committed in political actions (53% of men against the 47% of women; mean difference 1,47 against 1,14; sig. test,89), having taken at least one form of political action. The gender gap is significantly higher for collective actions rather than private ones (range of the variable: -1; mean for men in collective political activism 1,44 against 1,6 for women; in private political activism 3

4 mean for men 1,51 against 1,28 for women; Eta value in the difference of means tests respectively,1 and,5), especially for membership, leaflets distribution or attendance to political meetings. The gender gap disappears in the case of electoral voting, thus confirming the anomalous features of electoral participation compared to other forms of political participation. Tab. 3 Percentages of men and women involved in different political activities Women are not penalized in the case of private participation (e.g., boycott/buy products), where the individual initiative matters more, as other researches confirm (Coffé and Bolzendahl 21). Gender gap in interest for politics is high: 43% of men are interested in politics to a greater extent, against 28% of women (Eta value,192). This finding could reveal a lower interest among women toward traditional forms of political participation or national politics imaginary, but it doesn t exclude different attitudes toward local and small-scale political realities. The gender gap is also high in what concerns the competency level: knowledge about political institutions and political chronicles is significantly lower among women (19% against 1% of men; Eta value:,187). They also are more likely to feel uneasy when talking about politics (64% of women declare they are frequently or very frequently uneasy against 53% of men). The importance attributed to political parties is higher for men than women (range of the variable: -1; mean for men 5,7, mean for women 4,94; Eta value,28 in the difference of means tests). Tab. 4 Percentages of men and women who have taken at least a political action by principal socio-demographic characteristics. In a more detailed level of analysis, Table 4 confirms a reduction of the gender gap in political participation (especially in the collective dimension) within younger cohorts and it even disappears at top levels of education. The gap is high among retired (while it revers among unemployed), as well as among self-employed. In general, women display a stronger relationship between occupational status/class and participation. This descriptive framework proves our first hypothesis to be true. Women do participate less than men in political activities and are less interested in politics. They are less likely than men to take collective as well as private political actions and to be involved in visible as well as invisible political activities. Our second hypothesis holds also true: as political participation occurs, women prefer invisible rather than visible political involvement and private rather than collective political activism. 4

5 3. Men and women s participation and family constraints The hypothesis of familiar constraints requires a selection of the sample, including just women with young children. This information is directly estimated by the variable "age of children" in Istat survey, while it is only indirectly available from the Itanes data through a selection of women born after Moreover, selecting younger cohorts makes evident that gender gap is levelling off while the impact of parental constraints is more easily detectable. If scarcity of time due to familiar constraints does impact differently for men and women on political participation, it should have an effect on all forms of participation (visible as well as invisible, private as well as collective). Figures 1-2 presented in this section show the predicted probabilities of political participation calculated by taking into account the effects of socio-demographic covariates included in different regression models. 1 Fig. 1 Levels of political participation of men and women by marital status Figure 1 provides predicted probabilities of participating for men and women by marital status. Married people (both men and women) have a lower propensity of being politically involved, especially for visible forms of participation. The gender gap is higher for visible forms, but it is worth noting that married status seems to lower also men s political participation. In the case of political activism, people in other conditions than married or unmarried (divorced, separated, widowed) have a higher propensity to participate compared to other marital status. It is worth noting, here, that gender gap vanishes for private political activism, regardless of marital status. Fig. 2 Levels of political participation of men and women by number (and age) of children A far more important variable to the purposes of our analysis is the presence of children in the household. Our findings show some unexpected trends. Concerning visible forms of participation, we can see that having children does not affect exclusively women as a penalty in political involvement, controlling for other covariates (education, geopolitical zone, occupational status, and cohort). Among those born after 1965, we can see that men and women with two or more children have higher propensity to visible political involvement than those with just one child, and almost the same propensity as men and women with no children at all in the household. 2 1 We used logistic regression models and ran Margins procedure in Stata. Predicted probabilities correspond to proportions and can also be read as adjusted percentages. 2 All the analysis based on Itanes data and concerning the relationship between number of children and participation exclude unmarried women. This selection is necessary because the survey does not detect precisely the type of family 5

6 In other words, having more children does not lower the likelihood to participate, but leaves it unaltered, or even slightly higher compared to the situation of having one child. If at least one child (-5 years) is in the household, though, the propensity is lower, especially for women. Number of children does not affect invisible political involvement (getting informed, talking about politics and listen to political debates) both for men and for women. Itanes dataset provides a framework showing similar dynamics. Collective political activism is lower when just one child is in the household (compared to households with no children), but is even higher with two children, both for men and for women. Private political activism is higher, again, for people with two children, while it does not change for or 1 child. These U-shaped trends can be explained by referring to different dynamics arising in presence of children. Two contradictory images of having children are reported in literature: children as constraints and children as connections (Gallagher and Gerstel, 21). Children may be a constraint because the overall demand for domestic and care work increases (Mattingly and Bianchi 23), there is less free time for community networking and network maintenance (Munch et al 1997). Children can also be a resource because they make parents interact with more and diverse people and institutions (Moore 199; Nomaguchi and Milkie 23) and they spur participation in community and school life (Smith 1994; Wilson and Musick 1997) with a positive effect on politics when children enter the school-age (Voorpostel and Coffé 21). Our third hypothesis is thus only partly confirmed: family constraints play a crucial role in preventing people from participating both in political and non-political activities and this is true especially for women. Women s political involvement and activism result lower when they have more time-demanding roles (being married, having children, especially pre-school aged). Nevertheless, having two or more children could straighten their relationship with the social community and with political issues, thus mitigating the effect of scarcity of time. As for private activism this effect of children as connections is even greater, with a higher propensity of talking about politics, getting informed and listening to political debates. Interestingly, these dynamics are even emphasized for men. 4. Men and women s political interest and participation Istat survey provides information about various forms of non political participation: data about volunteering, being affiliated to trade-unions, ecology, peace and civil rights associations, cultural and recreational association, and professional associations. Apart from politics, the gender arrangement (the number of children for unmarried women could be computed also for women without their own children and refer to the composition of the family of origin). 6

7 gap is further confirmed for a formalized involvement across all non-political organizations and institutions, while it is smaller for other forms of social and cultural commitment. In any case, the overall involvement in this kind of activities regards small percentages of the sample: from a minimal commitment to ecology, peace and civil rights associations (1,9%) to cultural and recreational associations (1%). Among younger generations (those born after 1965), the gender gap is substantially maintained, except for cultural and recreational associations, where the gap is slightly reduced. An interesting test to assess our fourth hypothesis is comparing the relationship between family variables and political participation on the one hand, and family variables and other forms of participation on the other. One could assume that if a structural conditioning is at stake, the presence of children in the household and being married have the same impact in women s participation regardless of their political or non-political contents. Otherwise, if familiar constraints do impact only for political participation, then one would assume that the difference derives from a scarcity of interest towards political participation. In Fig.3 we have estimated predicted probabilities both for non-political participation (civic associations, cultural commitments, etc.) and leisure activities (going to the cinema, theatre, etc.) by marital status, number and age of children. Fig. 3 Levels of non political participation and leisure activity involvement of men and women by marital status, number and age of children At a first glance, we can see that women have a lower propensity to participate across all forms of non-political participation and leisure activities. We can note that: - The probability of being involved in social participation as well as taking part of leisure activities is lower for married than unmarried people, and this applies to both men and women. Women have a lower propensity than men in the same marital status to participate in all forms of non-political participation and involvement. - Having one child is the condition that leads to the lowest level of social participation. Here again, we can register a curvilinear trend, slightly ascending for people with two or more children. This trend is similar for women and men, but the penalty of having one child is stronger for women (especially with children -5 years old). As for the leisure activities very similar dynamics can be noted: a decline in the propensity to leisure activities in public for married people (but higher for separated / divorced, etc.), and a curvilinear trend with the number of children, with a decrease in the probability to perform leisure 7

8 activities when there are -5 years old children in the household. In all cases these trends are similar among women and men. In general, similarly to what emerges in fig. 2, structural conditioning is quite similar for men and for women regardless of the type of activity: being married signals a general reduction in the intensity of collective activities, whether political or non-political, even when controlling by socio-demographic characteristics. The number of children has a U-shaped trend that is higher for non-political participation and leisure activities. The age of children does matter, since having at least one -5 years old child reduces the propensity to participate. A final step in our analysis is an attempt to directly test the effect of motivation on political participation. If a cultural conditioning is at work, women with high political interest should not be affected by the negative effect of familiar constraints. That is, if more involved women s participation is not affected by structural conditions (having children and being married), it means that the gender differences previously highlighted are due (also) to the difference in interest, not just to the weight of familiar duties. This analysis is computed on Itanes dataset, since it provides information on interest in politics. As it does not provide information on age of children, we selected women born after 1965 (in order to deal with generally younger children) and excluded unmarried women (since the composition of their household, as asked in the survey, could include children who are not their sons or daughters). Fig. 4 Levels of political involvement of men and women by level of political interest The results in Fig. 4 suggest that in the same social conditions (cohort post1965, unmarried), both interested and non interested women with children have a lower propensity to participate to collective political activities (the relationship is negative and the proportion decreases as the number of children increases), while in case of individual activities women with one child are even more participatory than women with no children. However, results clearly show that political interest counts in general, regardless of family constraints. Different and gendered paths to political participation can explain women s weaker motivation to political participation: being socialized to feel uneasy with politics does not activate crucial resources to political activities. That is, when there is interest, women are as active as men (or even more active) no matter what the family constraints (marital status or number of children) are. Women are even more active when it comes to private and individual participation. As for collective activism, interested mothers have higher 8

9 propensity to participate (,635 for those with one child,,532 for those with two or more children) than the overall sample of childless women (,474). For private activism, interested mothers have higher propensity to participate (,72 for those with one child,,579 those with two or more children) than the overall sample of women with no children at all (,329). Our fourth hypothesis is thus partly confirmed: being socialized to feel extraneous to political arenas, women have lower interest and motivation in political participation. Nevertheless, when they are stimulated by political issues, interest boosts participation also for those women under the effect of family constraints. 5. Concluding remarks In our paper we investigate this issue moving from two different levels of analysis. One, more descriptive, aims at giving a general overview of several forms of participation of men and women, starting from socio-demographic differences. As a second point we deal with a more complex purpose. The key argument tested in the paper is the impact of family constraints on women s political participation, indirectly through marital status (married women less involved) or directly through an unbalanced division of time for childbearing that penalize mothers more than fathers. Results only partially confirm the relevance of this effect. Being married, as well as having a child, lower women s level of participation, but only with regard to visible political activities. Having a child in pre-scholar age is significantly depressive factor for participation but it affects only conventional political activities and mobilization. Results confirm that family constraints certainly play a role in preventing women from participating, especially when considering a collective, more institutionalized and pro-active participation (what we call the visible and mobilizing dimensions of politics). On the contrary, when it comes to less formalized and more individualized political activities (what we call the invisible dimension of participation), marital status and children do not affect participation by the same token. Our analysis reveals other interesting and even contradictory implications for the study of gender gap and family constraints. First, the effect of children follow an unexpected U-shape trend, with a significant rise in participation in households with more than one child. Women with two or more children have participation rates similar to women without children. One possible interpretation entails with new public and private arenas that opens up through children (especially when they are more than one). Parents start experiencing new place such as kindergartens, schools, baby recreational places that may require alternative forms of participation and involvement, and all 9

10 this help strengthen family connection into local community. A requirements for different time and effort s allocation within the family time budget might go in pair with a decrease in conventional and collective activities and an increase in the individual and private s. A second relevant aspect is the similar political participation s dynamic for men and women that share family characteristics (marital status and number of children). This is a very crucial point. A general interest for politics may draw parents attention from more conventional and public aspects of political participation to the more private and individualized. Thus, other dimensions namely lifestyles and consumption s choices become political through precise knowledge and action about production and consumption (buy/boycott), neighbourhood s life and activism, child school and other daily features. Similar dynamic for both men and women in private political participation could sustain the emergence of critical citizen (Norris 23). The paper aimed also at comparing political and non-political participative contexts as regards the relevance of gender gap and the effect of family constraints. As for the visible political participation married women register the lower level of involvement in the case both of social and leisure activities. Having one child (especially aged -5) is similarly negative, especially for social participation, while having more than one children relatively boost the propensity to be involved. What is surprising is that family constraints behave in the same way for men and women, reproducing the existents gaps. In sum, family characteristics simply show why people (not only women) participate less when family burden become harder, but not why women participate less than men. Lastly, we tested the impact of family constraints among women with different motivation towards politics (namely, different level of political interest). Data suggest that motivation counts: political participation rises sharply for the more interested women. Family constraints (children) do not seem to lower the effect of political interest. Interested mothers have higher propensity to participate than the overall sample of childless women. 1

11 References Amato, P. R. (2). The consequences of divorce for adults and children. Journal of Marriage and the Family, 62, Barbagli, M. & Maccelli, A. (1985), La partecipazione politica a Bologna, Bologna, Il Mulino. Barnes S.H. & Kaase, M. et al. (1979) Political Action. Mass Participation In Five Western Democracies. Beverly Hills, CA: SAGE. Bernstein, A. G. (25). Gendered characteristics of political engagement in college students. Sex Roles, 52, Burns, N. (27). Gender in the aggregate, gender in the individual, gender and political action. Politics & Gender, 3, Burns, N., Schlozman, K. L., & Verba, S. (1997). The public consequences of private inequality: family life and citizen participation. American Political Science Review, 91, Burns, N., Schlozman, K. L., & Verba, S. (21). The private roots of public action. Gender, Equality, and political participation. Cambridge: Harvard University Press. Chibber, P. (22). Why are Some Women Politically Active? The Household, Public Space, and Political Participation in India, in International Journal of Comparative Sociology 22 43: Clark, C. and Clark, J. (1986). Models of Gender and Political Participation in the United States. Women & Politics, 6, 1, Coffé, H., Bolzendahl C. (21). Same Game, Different Rules? Gender Differences in Political Participation. Sex Roles, 62, Conway, M., Steuernagel G., and Ahe, D. (24). Women And Political Participation: Cultural Change In The Political Arena. Washington, D.C.: CQ Press. Dalton, R.J. (1996). Citizen Politics: Public Opinion and Political Parties in Advanced Industrial Democracies. New York:Chatham House / Seven Bridges Press. Dow, J. K. (29). Gender differences in political knowledge: Distinguishing characteristics based and returns-based differences. Political Behavior 31(1), Fridkin, K., and Kenney, P. (27). Examining the gender gap in children s attitudes toward politics. Sex Roles, 56, Kalmijn, M. and Broese van Groenou, M. (25). Differential effects of divorce on social integration. Journal of Social and Personal Relationships, 22, Kingston, P. W. and Finkel, S. E. (1987). Is there a marriage gap in politics? Journal of Marriage and the Family, 49, Lister, R. (23). Citizenship: Feminist perspectives. New York: Palgrave Macmillan. Lovenduski, J. (1998). Gendering research in political science. Annual Review of Political Science, 1, Lovenduski, J. (25). Feminizing politics. Cambridge: Polity. Lowndes, V. (24). Getting On or Getting By? Women, Social Capital and Political Participation. The British Journal of Politics & International Relations, 6, Mattingly, M. J. and Bianchi, S. M. (23). Gender differences in the quantity and quality of free time: the U.S. experience. Social Forces, 81, Norris, P. (22). Democratic phoenix: Reinventing political activism. Cambridge: Cambridge University Press. Paxton, P., Kunovich, S., & Hughes, M. M. (27). Gender in politics. Annual Review of Sociology, 33, Rapoport, R. B. (1981). The sex gap in political persuading: where the structuring principle works. American Journal of Political Science, 25, Rotolo, T. (2). A time to join, a time to quit: the influence of life cycle transitions on voluntary association membership. Social Forces, 78,

12 Sapiro, V. (1982). Private costs of public commitments of public costs of private commitments? Family roles and political ambition. American Journal of Political Science, 25, Sayer, L. (25). Gender, time, and inequality: Trends in women s and men s paid work, unpaid work, and free time. Social Forces, 84, Schlozman, K. L., Burns, N., & Verba, S. (1994). Gender and the pathways to participation: the role of resources. The Journal of Politics, 56, Schlozman, K. L., Burns, N., & Verba, S. (1999). What happened at work today? : A multistage model of gender, employment, and political participation. The Journal of Politics, 61, Stake, J. (27). Predictors of change in feminist activism through women s and gender studies. Sex Roles, 57, Stoker, L. and Jennings, M. K. (1995). Life-cycle transitions and political participation: the case of marriage. American Political Science Review, 89, Stolle, D., Hooghe, M., & Micheletti, M. (25). Politics in the supermarket: political consumerism as a form of political participation. International Political Science Review, 26, Teorell, J.,Torcal, M. e Montero, J.R. (27) Political Participation:Mapping theterrain, in J.W.van Deth, J.R. Montero and A. Westholm (eds), Citizenship and Involvement in European Democracies: A Comparative Analysis. NewYork: Routledge, Verba, S., Burns N., and Schlozman K. (1997). Knowing and Caring about Politics: Gender and Political Engagement. The Journal of Politics, 59,

13 TABLES AND FIGURES Tab. 1 Percentage of people that engaged in political activities at least once in the last 12 months, by gender Men Women Tot Sig. test (Eta) Visible political involvement Conventional participation within parties Participation in a political party 5,2 2, 3,5,87 Donate or raise money for a party 2,9 1,6 2,2,43 Volunteering 1,8,7 1,3,51 Political mobilization Attended a rally 6,9 5,7 6,3,24 Attended a political meeting 7,2 3,8 5,4,75 Invisible political involvement Listening political debates 26,6 18,5 22,4,96 Getting Political Information Everyday 45,9 33,1 39,3 More than once a week 22,3 2,7 21,5 Once a week 4,1 3,8 3,9 More than once a month 7,5 8,6 8,1,169 More than once in a year 4,2 5,5 4,9 Never 16, 28,3 22,4 Talking about politics Everyday 16,2 9,1 12,5 More than once a week 3,1 2,7 25,2 Once a week 6,1 5,1 5,6 More than once a month 15,7 14,8 15,2,25 More than once in a year 9,6 1,6 1,1 Never 22,4 39,7 31,4 No political information: reasons* No interest 68,8 69,6 69,4 No time 6,5 5,7 6, Complex topic 9,6 14,7 13, No trust in politics 29,3 27, 27,8 Other 4,3 3,9 4, N** Note: *Multiple response. N=3731 (men) e 719 (women). ** N max. The analysis includes all men and women older than 14 years. Source: Istat

14 Tab. 2. Percentages of men and women politically involved by socio-demographic characteristics Visible political involvement (1) Invisible political involvement (2) Men Women Men Women In total 13,9 9, 45,8 34,9 Cohorts Pre-war 9, 3,6 59,8 36,2 Boomers 16,1 8,5 68,6 5,1 6s 15,3 9,7 6,6 48,5 7s 12,9 9,4 5,2 39,2 Post 7s 15,7 13,4 25,6 21,8 Sig. test (Eta),75,123,334,23 Education Low 11,1 6, 38,1 27,2 Medium 15,6 1,8 58,3 46,8 High 21,2 18, 77,2 61, Sig. test (Eta),97,138,268,254 Occupational condition Retired 1,5 4,9 62,2 44,3 Housewife - 4,7-33,2 Student 25,5 23,6 38,9 37,3 Unemployed 12,9 1, 39,5 34,8 Employed 14,1 12,2 55,5 47, Sig. test (Eta),111,186,159,134 Occupational class High 17,4 16, 74,1 63,2 Clerks 14,8 11,6 64,4 53,3 Self-employed 11,4 6,7 53, 39,4 Working class 1,9 5,5 44,8 34,4 Sig. test (Eta),76,127,229,29 Geopolitical zone North-West 11,9 8,4 48,5 39,7 Catholic zone 12,1 8,6 5, 41,1 Red Belt 12,4 8,2 51,2 4,4 Centre 15,7 1,1 47,1 37,2 South 16,4 9,6 37,8 23,8 Sig. test (Eta),59,24,17,153 N Note: (1) Visible: a) participation in a political party (yes); b) donate or raise money for a party (yes); c) volunteering (yes), d) attended a rally (yes); e) attended a political meetings (yes). (2) Invisible: getting political information (everyday); b) talk about politics (everyday); c) follow a political debate (yes). Source: Istat

15 Tab. 3 Percentages of men and women involved in different political activities Men Women Sig. test (Eta) Political actions (in general) % have done it 53,1 46,6 Means (-1) 1,47 1,14,89 Voted in 213 elections Yes 85,1 85,2 No 14,9 14,8,1 Collective political activism % have done it 43,6 35,6 Means (-1) 1,44 1,6,1 Attended a political meeting or debate 25,1 14,,136 Attended a rally 13, 8,6,77 Donate money for a party 3,7 2,1,66 Distributed leaflets 4,6 2,,81 Voted in primary elections 25,7 25,9,6 Private political activism % have done it 32,4 28,2 Means (-1) 1,51 1,28,5 Signed a petition 24,5 2,9,6 Send letters or claims to a public office 8,7 6,1,45 Boycott /bought products 12,3 11,3,2 Interest in politics To a great extent+somewhat 43,2 28,6 Very little+not at all 56,8 71,4,192 Political competence % no competence 1, 19,2 Mean (-1) 7,28 5,94,187 Perceived efficacy % no perceived efficacy 28,1 31,9 Mean (-1) 1,96 1,7,76 Unease when politics is debated Very frequently or frequently 52,8 63,7 Never, don t know, do not answer 47,2 36,3,116 Importance attributed to political parties* Mean (-1) 5,7 4,94,28 N Note: *The scale derives from three items: 1) political parties are needed, 2) thanks to parties people can participate, and 3) there is no democracy without parties. No perceived efficacy: minimum level in the scale -1. Source: Itanes

16 Tab. 4 Percentages of men and women who have done at least a political action by principal sociodemographic characteristics. Collective political actions (1) Private political actions (2) Men Women Men Women Total 43,6 35,6 32,4 28,2 Cohorts Pre-war 37,2 26,3 19,9 15,8 Boomers 39,9 24,5 36,2 22,4 6s 48,9 37,8 38,6 34,6 7s 44,8 36,5 29,8* 37,5 Post 7s 46, 46,5 35,3 28,1 Sig. test (Eta),84,174,141,164 Education Low 36,3 25,6 21,5 17,8 Medium 46,8 38,5 36,6 31,2 High 52,1 52,8 46,3 46,3 Sig. test (Eta),133,21,27,234 Occupational condition Retired 38,7 24,7 25,3 19,8 Housewife - 28,4 16, Student 57,1 53, 43,8 31,1 Unemployed 37,3 43,6 22,5 4,3 Employed 46,6 4,5 37,7 35,9 Sig. test (Eta),138,151,19,24 Occupational class High 46,8* 64,* 52,1* 5,* Clerks 56,8 43,4 5, 5, Self-employed 48,5 29,6 35,3 11,1 Working class 42,9 36,8 31,5 23, Sig. test (Eta),15,167,179,313 Geopolitical zone North-West 42,4 36,6 38, 34,4 Catholic zone 43,8 38,1 33,3 39,2 Red Belt 37,8 34, 28,9 25,3 Centre 4,3 28,9 37,2 23,1 South 5,8 38,6 25,4 22,1 Sig. test (Eta),92,7,111,141 N Note: N higher than 5 cases, except for * (1) Attended a political meeting or debate; Attended a rally; Donate money for a party; Distributed leaflets; Voted in primary elections. (2) Signed a petition; Send letters or claims to a public office; Boycott /bought products. Source: Itanes

17 Fig. 1 Levels of political participation of men and women by marital status Visible political involvement Invisible political involvement,2 M W,5 M W,4,15,3,1,2,5,1, Unmarried M arried Other cond. Unmarried M arried Other cond. Collective political activism Private political activism,4 M W,5 M W,4,3,3,2,2,1,1 Unmarried M arried Other cond. Unmarried M arried Other cond. Note: predicted probabilities controlling for other covariates (education, geopolitical zone, occupational status, and cohort). Cohorts born after Data computed using Margins. For the classification of visible vs. invisible participation and collective vs. private political actions see respectively tab. 2 and tab. 4. Source: Istat 211 (graph 1-2); Itanes 213 (graph 3-4). 17

18 Fig. 2 Levels of political participation of men and women by number (and age) of children Visible political involvement Invisible political involvement,2 M W,5 M W,15,4,1,3,2,5,1 No children 1 child 2+ children At least one child -5 No children 1 child 2+ children At least one child -5 Collective political activism Private political activism,4 M W,5 M W,4,3,3,2,2,1,1 No children 1 child 2+ children No children 1 child 2+ children Note: predicted probabilities controlling for other covariates (education, geopolitical zone, occupational status, and cohort). Cohorts born after Graphs do not include non married women. For the classification of visible vs. invisible participation and collective vs. private political actions see respectively tab. 2 and tab. 4. Source: Istat 211 (graph 1-2); Itanes 213 (graph 3-4). 18

19 Fig. 3 Levels of non political participation and leisure activity involvement of men and women by marital status, number and age of children Social participation,2 M W,9 Leisure activity involvement M W,15,8,1,7,5,6 Unmarried Married Other cond.,5 Unmarried Married Other cond. Social participation Leisure activity involvement,2 M W,9 M W,15,8,7,6,1,5,4,3,5,2,1 No children 1 child 2+ children At least one child -5 No children 1 child 2+ children At least one child -5 Note: predicted probabilities controlling for other covariates (education, geopolitical zone, occupational status, and cohort). Cohorts born after Data computed using Margins. Social participation: have participated in the activities of almost one of the following associations: tradeunions; volunteering associations; ecology, peace and civil rights associations; cultural and recreational associations; professional associations. Leisure activities: almost once in a year gone to: a) cinema; b) theatre; c) museum; d) music concert; e) sport event; f) disco; g) visit monuments. Source: Istat

20 Fig. 4 Levels of political involvement of men and women by level of political interest Collective political activism (men),8 M en (all) M en (interested),7,6,5,4,3,2,1,8,7,6,5,4,3,2,1 Private political activism (men) Men (all) Men (interested) Collective political activism (women),8 Women (all) Women (interested),7,6,5,4,3,2,1,8,7,6,5,4,3,2,1 Private political activism (women) Women (all) Women (interested) Note: predicted probabilities controlling for other covariates (education, geopolitical zone, occupational status, and cohort). Cohorts born after Data computed using Margins. Graphs do not include non married women. For the classification of collective vs. private political actions see tab. 4. Source: Itanes