The Gender Gap in political support in Palestine

Transcription

1 The Gender Gap in political support in Palestine Minna Cowper-Coles; ECPG Lausanne, June 2017 This paper shows there to be a gender gap in political support in the Palestinian Territories and then suggests and tests some explanations for that gap. The gender gap in political support is a difference in levels of support between men and women for the two main political 'parties' 1 in the Occupied Palestinian Territories, Fatah and Hamas. Although the gap has increased and decreased at different points, it has persisted in the polls for at least twenty years and shows women supporting Hamas more than men and Fatah less than men. The explanations for the gender gap include gender differences in socioeconomic status and religiosity. I find that when these factors are controlled for the gender gap becomes greatly reduced and is, in most polls, no longer statistically significant. The Gender Gap in political support in Palestine Data from opinion polls in the Occupied Palestinian Territories shows there to be a gender gap in political support. This, as yet unexplored, gender gap shows that women have, over the last twenty or so years, tended to support Hamas more than men, and have tended to support Fatah less than men. Figure 1 and Figure 2 show men's and women s percentage support for Hamas and Fatah from 1998 to The red dots indicate the percentage of women who responded that they support Fatah or Hamas, while the blue dots represent men's percentage support. These graphs show that there is a gender gap and it has persisted now for almost two decades. Figure 3 shows these two graphs in the format of a 'gender gap' graph. It uses a standard formulation of a gender gap (below) - altered to fit the Palestinian context - to show the gender difference in support for the two major political groups in Palestine (Jones, 2012). Palestine Gender Gap = ((% Women support Fatah - % Women support Hamas) - (% Men support Fatah - % Men support Hamas)) In Figure 3 the points further from zero indicate a greater gender gap, and those closer to zero indicate a smaller gender gap. In this graph, a negative value represents the percentage of women supporting Hamas being higher than the percentage of men supporting Hamas and the percentage of women supporting Fatah being lower than the percentage of men supporting Fatah. A positive result would indicate the opposite. While of interest in its own right this gender gap may reflect a wider trend in the region; indeed a quick look at the Arab Barometer suggests women in the Middle East more broadly may tend to favour Islamist parties more than men (Arab Barometer, ). 2 While acknowledging that the Occupied Palestinian Territories have their own particular political context, namely that of continued occupation, nonetheless an exploration of the causes of the gender gap in the Occupied Palestinian Territories may well be able to provide insights into a much more widespread phenomenon. This paper will first briefly discuss methods and data, then, drawing upon interview data and the academic literature, set out hypotheses for explaining the gender gap in political support in the Palestinian Territories. Finally, I will test these hypotheses using a logistic regression and draw some conclusions. 1 Fatah and Hamas both see themselves as political 'movements' rather than parties. As organisations with military wings they certainly defy standard definitions of political parties (Bröning, 2013). However, they have for several years played the role of 'political parties' in countless university elections, as well as municipal elections and the 2006 legislative elections. In the Palestinian Territories, the term 'hizb' or 'party' was used by many people to denote Hamas and Fatah as well as the term 'haraka' or 'movement'. 2 The Arab Barometer Third Wave shows a gender gap of -5.6, taken from responses to question 605a relating to preferences for a religious or non-religious political party, with women more likely than men to report support for a religious political party (49.4 percent compared to 47 percent) and less likely than men to report support for a non-religious political party (23.6 percent compared to 26.8 percent). 1

2 Mar 98 Jun 98 Nov 98 Jan 99 Jan 99 Apr 99 Jun 99 Jul 99 Sep 99 Oct 99 Dec 99 Jan 00 Feb 00 Jul 00 Jul 01 Dec 01 May 02 Aug 02 Nov 02 Apr 03 Jun 03 Oct 03 Dec 03 Mar 04 Jun 04 Sep 04 Dec 04 Mar 05 Jun 05 Sep 05 Dec 05 Mar 06 Jun 06 Sep 06 Dec 06 Mar 07 Jun 07 Sep 07 Dec 07 Mar 08 Jun 08 Aug 08 Dec 08 Mar 09 May 09 Aug 09 Dec 09 Mar 10 Jun 10 Oct 10 Dec 10 Mar 11 Jun 11 Sep 11 Dec 11 Mar 12 Jun 12 Sep 12 Dec 12 Mar 13 Jun 13 Sep 13 Dec 13 Feb 14 Jun 14 Aug 14 Sep 14 Dec 14 Mar 15 Jun 15 Sep 15 Dec 15 Mar 16 Jun 16 Sep Elections Hamas takes Gaza Operation Cast Lead Operation Pillar of Defense Operation Protective Edge Percentage support Support for Fatah according to Gender in the OPT over time Women who support Fatah Men who support Fatah Second Intifada Figure 1 Percentage support for Fatah by gender. Salient political events have been added to aid interpretation. Data=CPRS Polls 32, 33, 37-47, PSR Polls

3 Mar 98 Jun 98 Nov 98 Jan 99 Jan 99 Apr 99 Jun 99 Jul 99 Sep 99 Oct 99 Dec 99 Jan 00 Feb 00 Jul 00 Jul 01 Dec 01 May 02 Aug 02 Nov 02 Apr 03 Jun 03 Oct 03 Dec 03 Mar 04 Jun 04 Sep 04 Dec 04 Mar 05 Jun 05 Sep 05 Dec 05 Mar 06 Jun 06 Sep 06 Dec 06 Mar 07 Jun 07 Sep 07 Dec 07 Mar 08 Jun 08 Aug 08 Dec 08 Mar 09 May 09 Aug 09 Dec 09 Mar 10 Jun 10 Oct 10 Dec 10 Mar 11 Jun 11 Sep 11 Dec 11 Mar 12 Jun 12 Sep 12 Dec 12 Mar 13 Jun 13 Sep 13 Dec 13 Feb 14 Jun 14 Aug 14 Sep 14 Dec 14 Mar 15 Jun 15 Sep 15 Dec 15 Mar 16 Jun 16 Sep Elections Hamas takes Gaza Operation Cast Lead Operation Pillar of Defense Operation Protective Edge Percentage support Support for Hamas according to gender in the OPT over time Women who support Hamas Men who support Hamas Second Intifada Figure 2 Percentage support for Hamas by gender. Salient political events have been added to aid interpretation. Data= CPRS Polls 32, 33, 37-47, PSR Polls

4 Mar 98 Jun 98 Nov 98 Jan 99 Jan 99 Apr 99 Jun 99 Jul 99 Sep 99 Oct 99 Dec 99 Jan 00 Feb 00 Jul 00 Jul 01 Dec 01 May 02 Aug 02 Nov 02 Apr 03 Jun 03 Oct 03 Dec 03 Mar 04 Jun 04 Sep 04 Dec 04 Mar 05 Jun 05 Sep 05 Dec 05 Mar 06 Jun 06 Sep 06 Dec 06 Mar 07 Jun 07 Sep 07 Dec 07 Mar 08 Jun 08 Aug 08 Dec 08 Mar 09 May 09 Aug 09 Dec 09 Mar 10 Jun 10 Oct 10 Dec 10 Mar 11 Jun 11 Sep 11 Dec 11 Mar 12 Jun 12 Sep 12 Dec 12 Mar 13 Jun 13 Sep 13 Dec 13 Feb 14 Jun 14 Aug 14 Sep 14 Dec 14 Mar 15 Jun 15 Sep 15 Dec 15 Mar 16 Jun 16 Sep Elections Hamas takes Gaza Operation Cast Lead Operation Pillar of Defense Operation Protective Edge Gender gap Gender gap in political support for Fatah and Hamas in the OPT over time Second Intifada Figure 3 The gender gap in political support between Fatah and Hamas between 1998 and Data= CPRS Polls 32, 33, 37-47, PSR Polls

5 Methods This paper draws upon data from interviews with eighty-one people conducted between June 2014 and July 2016 with Palestinians from all sixteen governorates, in person or sometimes via Skype. Semi-structured interviews were carried out in English or in Arabic with an interpreter with interviewees who will remain anonymous. I also use opinion polls from the Palestinian Territories mostly conducted by the Palestinian Center for Policy and Survey Research (PSR) but also by its predecessor the CPRS. In total I have access to the response to the question Which party do you support? for seventy-five opinion polls (CPRS Polls 32, 33, 37-47, PSR Polls 1-61, including 52.1), but only have and use the full dataset from forty polls (PSR Polls 1-38, 42, 56). These forty PSR Polls were merged into a single SPSS file for ease of analysis. Further analysis was conducted using the software 'R'. Hypotheses There is a well-developed literature on gender and political support (and voting), which tends to centre on the gender gap in the United States, with fewer studies looking at other non- Western parts of the world. I use the academic literature together with what I was told in my interviews to unpick what might cause differences in political support between men and women. The academic literature suggests that women's political support may differ from men's due to gender differences in socio-economic status. Women are, in general, less well off than men on a global level (Inglehart, et al., 2003). Bergh explains: women have a lower average score on almost any measure of socioeconomic status, and, furthermore, that socioeconomic status generally correlates with voting' (Bergh, 2007 p. 237). In some cases, women's lower socioeconomic status is found to partially explain why they tend to vote more for 'liberal' or 'left of center' parties (Bergh, 2007 p. 237; Howell, et al., 2000 p. 860; Studlar, et al., 1998; Box- Steffensmeier, et al., 2004). That socioeconomic status might account for the gender gap in political support also came through in my interviews. The economy, and unemployment was a matter of concern to both the men and women I interviewed. The Palestinian Central Bureau of Statistics reported that the unemployment rate in Gaza was 41.2 percent in the Gaza Strip in the first quarter of 2016, and 18 percent in the West Bank and that the average monthly salary was 2250 NIS (the equivalent of approximately 600 US Dollars) (Palestinian Central Bureau of Statistics). Economic pressures (and the related drivers of political support) are particularly gendered. In the Palestinian Territories, the responsibility of 'providing' for family, i.e. financially looking after parents, siblings, wives or children, largely falls on men (Baxter, 2007; Gren, 2015 p. 155). Interviewees told me that men are not just expected to financially support the family, but they should also build a house and when they get married they are expected to pay for all the costs of the wedding which are huge (M13, 2014; 22, 2016; 23, 2016; 49, 2016). Many of the young men I spoke to felt that they would never be able to afford to get married or maybe in a hundred years (49, 2016; 40, 2016; 66, 2016). Those feeling financial pressure, in a context of high levels of unemployment and low salaries, might be inclined towards supporting Fatah, a party which is seen to support the peace process and economic development, giving men in a volatile region a greater ability to provide for their families. 5

6 It is not just Fatah s economic and peace policies that mean it might be a wiser choice for those who need to provide for their families, but Fatah membership or loyalty to the party can provide greater access to work opportunities. One interviewee told me: It s a rule All the people who take good jobs at a high level must be [with] the government... If you are Fatah they give you support to get the job, and to go to up. (50, 2016) Another interviewee saw the benefits that Fatah provided as particularly gendered. She said: Fatah gives this sustainable constant work you can get privileges from working with Fatah just following them opens doors for you, whether it is in your own family or in your workplace... I think women don t need that. The pressure on men to sustain... their status and their family is more than women. (35, 2016) If being associated with Fatah opens doors (35, 2016), Hamas, on the other hand, might be considered as being an economic 'risk'. The international community (apart from perhaps certain Gulf states and Iran) refuses to provide aid to Hamas and its associated organisations (65, 2016; Caridi, 2012 p. 196; Shamir, et al., 2010 p. 7). When Hamas won the 2006 election international aid was instantly blocked. Since they took over the Gaza Strip in 2007, Israel has blockaded Gaza ruining its economy. An interviewee told me: Hamas was seen as the number one reason why there are no jobs in Gaza and men are usually the ones who are responsible [for providing for the family] (67, 2016) In the West Bank as well, Hamas supporters will find obstacles and problems, with his situation, his future (37, 2016). One interviewee told me: if you support Hamas... I go to take a job and they... refuse me. (50, 2016). As such we might suppose that more men than women will support Fatah in order to gain or keep employment, and might not support Hamas because of the risk they pose to economic stability. Women do not face the same pressure to 'provide' as men do, but their socioeconomic status may affect their political support in a different way. Women tend to have fewer economic resources than men, particularly when women are unmarried, divorced or widowed (Miftah, 2013; Palestinian Central Bureau of Statistics, 2014). Women also play the main caring role for children, the sick and the elderly, and in this role are likely to benefit or come into contact with charities, and the welfare services more than men. For these reasons, women might be influenced by political parties through their distribution of services and aid. Hamas is well known for its distribution of services as Sara Roy describes in her book (Roy, [2011] 2014). Hamas in Gaza has been known to help with food, university fees wheelchairs, furniture or clothes, gas, kitchen accessories and of course medical assistance also (60, 2016). Access to these welfare services is usually through women. One interviewee told me that in Gaza: The economic situation is very bad and that means there is no work and so the result is they get aid from institutions. And men find it very difficult to go to institutions for their family and ask for money. So that is the women s role now. she goes to those institutions and asks for money and she votes for those institutions that provide her with money. (62, 2016) This suggests that women might support Hamas more than men because they are more closely linked to and more likely to be dependent on the welfare and support they provide. 6

7 It is of course worth noting that the role of the family, where economic fortunes will be shared by the group, can be expected to moderate gender differences. However, there are clear theoretical justifications for men and women being subjected do different financial pressures and thus relating to political groups differently in the Palestinian Territories. As such we might expect that the gender gap in socioeconomic status accounts for part of the gender gap in political support. Thus, the first hypothesis to be tested is: Hypothesis 1: Gender differences in socioeconomic status partially account for the gender gap in political support. A second prominent feature of the academic literature in accounting for a gender gap in political support is religiosity. Religion is often also intimately linked to gender ideals, as well as questions of identity and nationalism (Kandiyoti, 1991; Yuval-Davis, 1997/2010). In the Occupied Palestinian Territories, where the struggle against occupation persists, nationalism and religion play an important part in politics. Indeed, the role of religion and religiosity in determining political support is likely to be highly relevant as it is one of the major distinctions between the largely secular Fatah and Islamist Hamas. Several academics, consider religiosity as a major factor impacting how men and women vote, and have often found women to be more religious than men (Howell, et al., 2000 p. 869; Norrander, et al., 2008 p. 521; Inglehart, et al., 2003; Edlund, et al., 2002; Desposato, et al., 2008). When I asked interviewees what might explain the gender gap, the primary explanation that was put to me was that women are more religious than men, and so they are more likely to support the Islamist resistance movement, Hamas (M1, 2014; M5, 2014; M12, 2014; M13, 2014; 7, 2015; 15, 2015; 20, 2016; 28, 2016; 31, 2016; 34, 2016, 37, 2016; 41, 2016; 52, 2016; 65, 2016; 68, 2016). Interviewees provided numerous explanations as to why women might be more religious than men. While faith and personal belief should be readily accessible to both sexes, a gendered social context might encourage religious behaviour more in one sex more than the other. Women and girls, I was told, were held to higher moral standards than men and boys (20, 2016; 21, 2016; 31, 2016; 43, 2016). Staying out late, smoking and drinking were behaviours that were forgivable in a man (perhaps) but were much more problematic for a woman. I was told that a woman who behaved like this was seen as the tramp of the village and that it definitely affects her long term (20, 2016). And if a woman was known to have been drunk say - it stays with her for the rest of her life. No one will marry her (31, 2016). Women s morality is watched and safeguarded more than men s in Palestinian society and women risk losing greater status by breaching these norms (see Baxter, 2007). Theory suggests that nationalism also indirectly increases women s religiosity, by increasing the pressure on women to behave in certain ways. The nationalist project in Palestine dominates the political and social sphere. What theorists (such as Nira Yuval Davis) point out is that these national projects, by searching for cultural authenticity, or emphasising religious behaviour, often force women into traditional, apparently authentic, female roles, and burden them with a greater responsibility for behaving in certain moral ways (Yuval- Davis, 1997/2010; Kandiyoti, 1991). Two women interviewees (among many others) explained the importance of religion and culture to the national project: we have this belief that our Palestine will be free if we believe in the right Islam and we achieve good values (M1, 2014) and another told me If we lose part of the culture, we could be losing part of 7

8 being Palestinian (55, 2016). In the Palestinian Territories, where culture and religion are privileged parts of the national project, the implications of non-conformity are much higher. To break away from social expectations is, somehow, to fracture or betray the national cause. This again reinforces the pressures on women to conform to social norms of behaviour, and may explain the greater religiosity of women. One aspect of this social conformity involves the regulation of spaces that women can occupy. Certain public spaces, work spaces and even virtual spaces are often unfriendly for women. One female interviewee told me that the streets are male dominated (22, 2016) and another was told when applying for a job at a bank, "You can t come and work here because as a female you should be at home" (44, 2016). Women also have their online behaviour limited and subjected to double standards. A young woman I spoke to told me On Facebook a guy can have both female and male friends while the girl shouldn t have male friends. (43, 2016). The occupation has meant that women s mobility is even more limited from fear of attack from settlers, the harassment from soldiers at checkpoints and because if the worst comes to the worst and checkpoints are closed, it is easier and more socially acceptable for men to stay at a friend s house or walk through the fields to get home. The restrictions on women s work and movement might have given women more 'opportunity' to become religious or cultivate their moral standing. Several interviewees suggested that perhaps women have the time for religion from being at home. One interviewee told me they don't have work or other things to do so they become more religious (M5, 2014). Another said: women in general are given the role of staying at home so they have that free time to focus on more spiritual things whereas men are always out or hanging out or working women explore religion because they have the time for it, whereas men can occupy themselves with either going to smoke shisha or going to play cards or something like that, whereas a woman she can t do that. (20, 2016) The idea that women have more time reflects the fact that women have less space due to the social restrictions placed upon women s movements and activities. A third interviewee told me: The men in Gaza they go out to meet, they go to the coffee shop, they are influenced by other opinions and the women are not women in Gaza are more conservative, so they are closer to the mosque, they are closer to the social network. Men can go out, talk to somebody else, travel, or they go to their work in society, participate in the activities of civil society so they are open to other ideas. (60, 2016) Another implication of women s restricted access to certain areas of public space is that women, in spending more time at home, might be consuming more Islamic media on television and on the radio (17, 2015; M10, 2014; 68, 2016). While women, like those described above, who conform to traditional expectations might embrace religion, also several of those who criticise tradition because of the restrictions it places upon women, do so in the name of religion. These interviewees did not use the term feminism, they instead spoke in terms of women's rights. One interviewee told me: Religion gives women every single right that they deserve but society never gives them a right (26, 2016). I was given many examples of the much greater rights accorded to women in Islam 8

9 compared to the rights traditionally granted them. I was told that while traditionally society disapproved of women working, Islam permitted it (42, 2016; 43, 2016); tradition said a woman caught behaving inappropriately with a man should be killed, but Islam suggested a lesser punishment (M10, 2014); and while traditionally women had no inheritance, Islamic law allocates women half that of their brothers (40, 2016; 51, 2016; 62, 2016). From this perspective, it is in women s interests to be religious, as a religious woman they can claim more rights and freedoms for themselves than they might be permitted otherwise. Scholars such as Lihi Ben Shitrit and Saba Mahmood explore this alternative understanding of women's religious devotion (Mahmood, 2005; Shitrit, 2013). Ben Shitrit in her study of Jewish and Islamic movements in Israel, finds that for many women religious movements offer real liberation from oppressive socioeconomic realities and limiting cultural norms (Shitrit, 2013 p. 81). In the same way that religious rhetoric can be used to gain greater rights, being part of an Islamist nationalist organisation can provide rewards for women. Indeed, Hamas was seen by some as more accessible than secular organisations for women. Several of those I interviewed told me that it was easier to be politically active within Hamas without challenging social norms. A female interviewee told me: [as] a woman I would think that being part of Hamas, which is more conservative, would be better for me It would give me this look of [a] polite, quiet, religious person. And my family agrees with it because I would not have to go out a lot, I would be doing nothing wrong because they are religious." (25, 2016) Within Hamas, there is a women s section which is mostly separate from the male structure - I was told it is like a different political party (68, 2016; 42, 2016). Women work alongside and amongst other women in Hamas. Because of the parallel gender structure, many interviewees felt that women were given a bigger voice and role within Hamas than in other political parties. An interviewee told me: I think for a religious woman who still wants to maintain her religion as part of their political activism then yes they would choose that because they have a lot of power, and at the same time they would have a good platform and ideology (22, 2016) Hamas therefore might provide women with an organisation that values them and their work, and provides them with roles that do not confront social expectations. Scholar Maha Azzam understands this to be the case for some women. She writes: "under the umbrella of Islam, women may find it easier to demand a role, even though it may not be that of a judge or head of state, so long as they are functioning within strict rules both on the level of appearance and in voicing Islamic principles." (Azzam, 1996) Based on the literature and the explanations offered in the interviews, I suggest testing a second hypothesis that gender differences in religiosity accounts for (at least to some extent) the gender gap in political support. Hypothesis 2: Gender differences in religiosity account for part of the gender gap in political support. 9

10 Other hypotheses could be drawn from the academic literature concerning gender and political support, relating to support for feminist principles or support for non-violence (Conover, 1988; Bergh, 2007; Ruddick, 1989; Gilligan, 1982). Or else another hypothesis could be drawn from the interviews, that men risk greater 'retaliation' from the security services for supporting Hamas than women. While these theoretical approaches certainly warrant greater exploration, they are not considered in more detail here. These factors are likely to be complex and problematic because the academic literature and interview data are not consistent on any of these points, thus the cause of any finding might be unclear. In addition, testing these hypotheses would be more problematic as there are no clear questions relating to these factors throughout all of the polls. As such this paper focuses on the two above hypotheses relating to socioeconomic status and religiosity. Findings In order to test the two hypotheses set out above, I conducted logistic regressions on the two dependent variables, support for Hamas and support for Fatah using five different models, including up to ten variables. The graph displaying the gender gap in predicted probability of voting Hamas and Fatah (using the equation above) for each of the different models is shown below (Figure 4). Below are the different models. An explanation of the variables can be found in Appendix 1. Model 1: Fatah support ~ Gender Hamas support ~ Gender Model 2: Fatah support ~ Gender + Age + GazaDummy + RefugeeCampDummy + CityDummy Hamas support ~ Gender + Age + GazaDummy + RefugeeCampDummy + CityDummy Model 3: Fatah support ~ Gender + Age + GazaDummy + RefugeeCampDummy + CityDummy + Income + Employment + Education Hamas support ~ Gender + Age + GazaDummy + RefugeeCampDummy + CityDummy + Income + Employment + Education Model 4: Fatah support ~ Gender + Age + GazaDummy + RefugeeCampDummy + CityDummy + Income + Employment + Education + Religiosity Hamas support ~ Gender + Age + GazaDummy + RefugeeCampDummy + CityDummy + Income + Employment + Education + Religiosity 10

11 I would expect Models 1 and 2 to have very similar gender gap in predicted probability of supporting Hamas or Fatah. The region and age of respondents are being included in order to control for these factors which may impact political support. Changes in the political context is taken into account by conducting several logistic regressions for the different polls over a fifteen year period. By including income, education and employment in Model 3, I would expect it to produce a smaller gender gap in predicted probability compared to Models 1 and 2, because by holding socio-economic status the same for men and women the theoretical and interview accounts suggest that men and women might have less reason to support different political parties. This would disprove the null hypothesis for Hypothesis 1. I expect Model 4 to produce a reduced gender gap in predicted probability compared to Models 1, 2 and 3, because the gender differences in religiosity are likely to account for part of the gender gap in support for Hamas and Fatah. This will disprove the null hypothesis for Hypothesis 2, indicating that gender differences in religiosity do partially account for the gender gap in political support. 11

12 Jul 00 Jul 01 Dec 01 May 02 Aug 02 Nov 02 Apr 03 Jun 03 Oct 03 Dec 03 Mar 04 Jun 04 Sep 04 Dec 04 Mar 05 Jun 05 Sep 05 Dec 05 Mar 06 Jun 06 Sep 06 Dec 06 Mar 07 Jun 07 Sep 07 Dec 07 Mar 08 Jun 08 Aug 08 Dec 08 Mar 09 May 09 Aug 09 Dec 09 Mar 10 Jun 10 Oct 10 Dec 10 Dec 11 Jun 15 Gender Gap M1 (Gender) M2 (M1 + Age + Region) M3 (M2 + Income + Employment + Education) M4 (M3 + Religiosity) Figure 4 Graph to show the gender gap in predicted probabilities of men and women supporting Hamas and Fatah for Models 1-4. Data = PSR Polls 1-38, 42, 56 12

13 Figure 4 shows that, as expectedm, Models 1 and 2 produce a very similar gender gap in predicted probability of political support. Model 3 produces a reduction overall in the gender gap, but the impact of education, income and employment changes across time. This may be because at fraught times other factors were more salient for Palestinians than the economy. Nonetheless, we can accept Hypothesis 1 that socio-economic status does in fact partially account for the gender gap in political support. We might speculate that if employment and income were more evenly distributed between men and women in the Occupied Palestinian Territories there would be a smaller gender gap. Model 4 shows that when religiosity is held constant there is a smaller gender gap in predicted probability of political support than in Models 1, 2 or 3. As such we can accept that the gender differences in religiosity account for part of the gender gap in political support, thus accepting Hypothesis 2. It should be noted that in the first poll (July 2000) there were no variables for religiosity which explains why Model 4 and Model 3 produce similar results here. Overall, when socioeconomic status and religiosity are held constant, the gender gap is reduced by over In many instances the gender gap is halved or removed completely. Figure 4 does, however, show that the gender gap still exists even once socio-economic status and religiosity have been accounted for. Some alternative factors were suggested in the literature, such as gender differences in feminist values or attitudes to war and peace leading to gender gaps in political support. In the interviews it was also suggested that it is more dangerous for men to openly support Hamas in the West Bank. These factors would require further exploration, which is beyond the scope (and word count) of this paper. It is worth, however, including two final graphs, to show the statistical significance of the gender coefficient in each of the polls. Figures 5 and 6 show a comparison of the gender coefficients of Model 1 and Model 4 for support for Fatah and support for Hamas. The statistical significance of these coefficients has been added to the graph in the form of solid, dashed and dotted lines. Where these lines 'cross' the zero line, the gender coefficient can no longer be seen to be statistically significant at that level. As such, these two graphs show that while the gender coefficient is mostly statistically significant for each of the 40 polls being examined in Model 1, it is mostly not statistically significant in Model 4. As such the gender gap the pattern cannot be definitely distinguished from random chance once socioeconomic factors and religiosity are accounted for. 13

14 Figure 5 Graph to compare the regression coefficients of gender on support for Fatah with levels of significance denoted through solid (p<0.05), dashed (p<0.01), and dotted (p<0.001) lines for Model 1 and Model 4. Data= PSR Polls 1-38, 42, 56 14

15 Figure 6 Graph to show the regression coefficients of gender on support for Hamas with levels of significance denoted through solid (p<0.05), dashed (p<0.01), and dotted (p<0.001) lines for Model 1 and Model 4. Data= PSR Polls 1-38, 42, 56 Conclusion This paper has both shown there to be a gender gap in political support in the Occupied Palestinian Territories, and gone most of the way towards explaining it. It seems that gender differences in socioeconomic status and religiosity explain why men have tended to support Fatah more than women and women have tended to support Hamas more than men. The explanations of socio-economic status and religiosity will be familiar to those who study the gender gap. This analysis shows that political scientists are right to continue including and investigating the relationship between gender, socio-economic status, religiosity and voting preference, as they have a widespread power of explanation. This paper has contributed to a different understanding of women's greater religiosity, and questioned the view of women's religiosity as something straightforward and 'traditional'. This paper has also highlighted issues normally outside of the remit of analyses of polling data and gender differences. Here I have touched upon the gendered effects of nationalism on political support and the role of nepotism in swaying men towards the status quo. While the main explanations of the gender gap are not so different from what we might see in the West, there is a major role played by the political context of the occupation and nationalism in explaining the gender gap. The Israeli Occupation places additional pressures upon the economy and makes certain political behaviours riskier and these pressures have played a role in accentuating gender differences. Nationalist rhetoric has increased pressure on women for social conformity and in this context Islamist nationalism has become a socially 15

16 acceptable method for participation in nationalist activities for women. These insights should encourage a more nuanced understanding of the interplay between gender and political support in the wider Middle East and North Africa, or in post-colonial political contexts. Interviews: M1 - Nablus Governorate: 14 July younger, female, in Arabic with interpreter. M5 - Nablus Governorate: 15 July younger, male, in English. M10 - Nablus Governorate: 23 July middle-aged, male, in English. M12 - Nablus Governorate: 24 July middle-aged, male, in Arabic with interpreter. M13 - London/ Nablus Governorate: July younger, female, by Skype, in English. 7 (Expert) - Ramallah Governorate: 28 May middle-aged, male, in English. 15 (Expert) - Ramallah Governorate: 14 June middle-aged, male, in Arabic with interpreter. 17 (Expert)- Ramallah Governorate: 24 June middle-aged, female, in English London: 14 January younger, female, in English London: 22 January younger, female, in English London: 24 January younger, female, in English London: 27 January younger, male, in English Ramallah Governorate: 25 February younger, female, in English Jerusalem area: 25 February older, female, in Arabic with interpreter. 28 (Expert) - Jenin Governorate: 28 February middle-aged, female, in Arabic with interpreter Jerusalem area: 29 February younger, female, in English Ramallah Governorate: 2 March middle-aged, male, in Arabic with interpreter Jericho Governorate: 3 March younger, female, in English Toubas Governorate: 6 March younger, male, in Arabic with interpreter Nablus Governorate: 6 March younger, male, in English. 41 (Expert) - Ramallah Governorate: 7 March middle-aged, female, in English 42 (Expert) - Ramallah Governorate: 8 March older, male, in Arabic with interpreter Ramallah Governorate: 8 March younger, female, in Arabic with interpreter, with interviewee Ramallah Governorate: 8 March younger, female, in Arabic with interpreter with interviewee Hebron Governorate: 13 March younger, male in English. 16

17 50 - Hebron Governorate: 13 March younger, male in English and Arabic. 51 (Expert) - Nablus Governorate: 14 March older, female, in Arabic with interpreter. 52 (Expert) -Ramallah Governorate: 16 March older, male, in English Bethlehem Governorate: 23 March younger, female, in English. 60 (Expert) - Gaza/London: 27 April middle aged, male, via Skype, in English Gaza/London: 4 May middle aged, female, via Skype, in Arabic with interpreter Sussex: 16 May younger, female, in English London: 11 May younger, male, in English London: 19 May younger, female, in English London: 29 July younger, male, in English. Bibliography Arab Barometer Arab Barometer Third Wave [Online] // Arab Barometer Azzam Maha Chapter Nine: Gender and the Politics of Religion in the Middle East [Book Section] // Feminism and Islam: Legal and Literary Perspectives / book auth. (Ed) Mai Yamani. - Reading : Ithaca Press, Baxter Diane Honor Thy Sister: Selfhood, Gender, and Agency in Palestinian Culture [Journal] // Anthropological Quarterly : Vol pp Bergh Johannes Explaining the Gender Gap: A Cross National Analysis of Gender Differences in Voting [Journal] // Journal of Elections : Vol pp Box-Steffensmeier Janet M., De Boef Suzanna and Lin Tse-min The Dynamics of the Partisan Gender Gap [Journal] // The American Political Science Review. - August : Vol pp Bröning Michael Political Parties in Palestine: Leadership and Thought [Book]. - New York : Palgrave Macmillan, Caridi Paola Hamas: From Resistance to Government [Book] / trans. Teti Andrea. - New York : Seven Stories Press, Conover Pamela Johnston Feminists and the Gender Gap [Journal] // Journal of Politics : Vol pp Desposato Scott and Norrander Barbara The Gender Gap in Latin America: Contextual and Individual Influences on Gender and Political Participation [Journal] // British Journal of Political Science Vol pp Edlund Lena and Pande Rohini Why Have Women Become Left-Wing? The Political Gender Gap and the Decline in Marriage [Journal] // The Quarterly Journal of Economics. - August : Vol pp Gilligan Carole In a Different Voice: Psychological Theory and Women's Development [Book]. - Cambridge Massachusetts: Harvard University Press,

18 Gren Nina Occupied Lives: Maintaining Integrity in a Palestinian Refugee Camp in the West Bank [Book]. - Cairo : American University Cairo Press, Howell Susan E. and Day Christine L. Complexities of the Gender Gap [Journal] // The Journal of Politics : Vol pp Inglehart Ronald and Norris Pippa Rising Tide: Gender Equality and Cultural Change Around the World, Ronald Inglehart and Pippa Norris [Book]. - Cambridge : Cambridge University Press, Jones Jeffrey M. Gender Gap in 2012 Vote Is Largest in Gallup's History [Online] // Gallup. - 9 November February vote-largest-galluphistory.aspx?g_source=gender%20gap&g_medium=search&g_campaign=tiles. Kandiyoti Deniz Identity and its Discontents: Women and the Nation [Journal] // Millenium - Journal of International Studies pp Mahmood Saba Politics of Piety [Book]. - Princeton : Princeton University Press, Miftah The Palestinian Initiative for the Promotion of Global Dialogue and Democracy Gender-based analysis of the services provided by the Ministry of Social Affairs: A beneficiary-based Study [Report] Norrander Barbara and Wilcox Clyde The Gender Gap in Ideology [Journal] // Political Behavior. - December : Vol pp Palestinian Central Bureau of Statistics [Online] &wversion=Staging. Palestinian Central Bureau of Statistics Women and Men in Palestine: Issues and Statistics [Report]. - Ramallah : [s.n.], Roy Sara Hamas and Civil Society in Gaza: Engaging the Islamist Social Sector [Book]. - Princeton : Princeton University Press, [2011] Second Edition. Ruddick Sara Maternal Thinking: Towards a Politics of Peace [Book]. - Boston Massachusetts: Beacon Press, 1989 Shamir Jacob and Shikaki Khalil Palestinian and Israeli Public Opinion: The Public Imperative in the Second Intifada [Book]. - Bloomington and Indianapolis : Indiana University Press, Shitrit Lihi Ben Women, Freedom, and Agency in Religious Political Movements: Reflections from Women Activists in Shas and the Islamic Movement in Israel [Journal] // Journal of Middle East Women's Studies : Vol pp Studlar Donley T., McAllister Ian and Hayes Bernadette C. Explaining the Gender Gap in Voting: A Cross-National Analysis [Journal] // Social Science Quarterly. - December : Vol pp Yuval-Davis Nira Gender and Nation [Book]. - London : Sage, 1997/

19 Appendix 1 Dependent Variables: Fatah support: A dummy variable where 0 = does not support Fatah and 1 = supports Fatah Hamas support: A dummy variable where 0 = does not support Hamas and 1 = supports Hamas These variables were recoded from a 'Which party or movement do you support?' question where respondents were given a choice of thirteen or so options (the exact number varies from poll to poll as smaller parties were added to the choices) including Hamas, Fatah, several leftist and smaller parties, independents and not supporting any party. Independent Variables Gender: A dummy variable where 0 = male and 1 = female. Age: Age in years Gazadummy: A dummy variable where 0 = does not live in Gaza and 1 = lives in Gaza Refugeecampdummy: A dummy variable where 0 = does not live in a refugee camp and 1 = lives in a refugee camp Citydummy: A dummy variable where 0 = does not live in a city and 1 = lives in a city IncomeSummary: A three-point scale to indicate higher and lower reported family/household income levels, This variable combines three different income variables from different polls and has recoded them onto a three-point scale. Employment: A dummy variable where 0 = those not employed in the public or private sector and 1 = those employed in the public or private sector. Education: A seven-point scale to indicate the highest level of education reached. Here the levels and coding are as follows: 1 = illiterate, 2 = elementary, 3 = preparatory, 4 = secondary, 5 = college diploma, 6 = bachelor s degree, 7 = master s degree and above. Religiosity: A four-point scale where 0 = not at all religious, 4 = religious. This variable combines three different measures of religiosity, self-identification as religious, frequency of prayer and frequency of reading the Qur'an, each of which are only present in certain polls. Unfortunately none of these measures of religiosity were included in Poll 1. 19