Female University Students Respond to Gender Violence through Dialogic Feminist Gatherings

Within the framework of the “ Free_Teen_Desire” research project led by the University of Cambridge and funded by the programme Marie Curie Actions 1 , a survey was conducted. Vignette-Test data for 127 female university students (ages 18-27 years) in Spain reveals that the wish to hook up with a violent young man significantly decreases after a gathering on the topic of the Mirage of Upward Mobility, a successful programme elaborated in Dialogic Feminism (Beck-Gernsheim, Butler & Puigvert, 2003). In the pre-test, 78.4% of the respondents stated that their female friends would like to hook up with a violent man at a party, while this percentage decreased to 38.5% when they responded concerning themselves. After the pre-test, there was a one-hour gathering and debate. The subsequent post-test revealed that only 48.8% of the respondents stated that their female friends would like to hook up with a violent man at a party, and 14.9% of the respondents made the same statement concerning themselves. The survey presented pictures of four men accompanied by a short explanation of their characters. The explanations of man 1 and man 3 included sentences that describe behaviours characterized as gender violence in previous international surveys (Banyard et al., 2005; Fisher, Cullen, & Turner, 1999; Gross et al., 2006; Kalof et al., 2001). The descriptions of man 2 and man 4 only included non-sexist behaviours. The data did not significantly change when we exchanged the pictures of man 2 and man 4 in the instrument with the pictures of the men with violent profiles and then administered the post-test. For different groups of respondents, the period between the pre-test and the gathering as well as between the gathering and the post-test were changed from fifteen minutes to one and two weeks. In all cases, we obtained similar results for the pre- and post-tests. However, additional research is required to demonstrate how long the effect of the gathering endures and to identify the processes that can increase or decrease the effect over time.

ender violence among young women is a worldwide scourge. Nearly one of four girls aged 15 to 19 years has experienced some form of physical violence (UNICEF, 2014). In the United States, Taylor and Mumford (2016) used data from the National Survey on Teen Relationships and Intimate Violence and demonstrated that nearly one of five adolescents aged 12 to 18 years reported being victims of sexual and physical abuse. In a multi-country study by Decker et al. (2014) on female adolescents aged 15-19 years, the authors demonstrated that 18.6% of adolescents in Johannesburg, South Africa, have been pushed or shoved. In the study, the authors considered intimate partner violence and non-partner violence. In Europe, in an EU-wide survey, 34% of women respondents under 29 years reported having suffered physical and/or sexual violence at the hands of a partner or a non-partner since the age of 15 (FRA, 2014). The UK results are similar: 36% of the rapes registered by the police were perpetrated against girls under the age of 16 years (British Office for National Statistics, 2010). In Spain, a pioneering study demonstrated that 62% of the students know of or have experienced situations of gender violence in the university context although only 13% recognized these situations as such (Valls, Puigert, Melgar & García-Yeste, 2016). However, less is known regarding the past relationships of these students during secondary education and the effectiveness of the programmes in which they have participated with respect to identifying and preventing gender violence in their lives.
The design and development of effective measures, actions, programmes and interventions to contribute to overcoming gender violence are at the core of the international research agenda. To support the potentially useful impact of these initiatives, it is important to emphasize the relevance of basing their implementation on research evidence. Accordingly, emerging insights regarding additional social and political interventions could be enhanced. This approach is in line with an increasing concern among scientists regarding how to achieve a positive impact as a result of research (Flecha, Soler-Gallart & Sordé, 2015). Therefore, additional research is required to increase our knowledge regarding the type of programme that can contribute to overcoming gender violence among young women. To prevent gender G 186 Puigvert -Female Univesity Students Respond to Gender Violence violence in schools, universities, leisure places, homes and neighbourhoods, such research is urgently required.
To this end, this article examines the effect of a Feminist Gathering on female university students. First, the state of the research on the attraction to violence and the prevention of gender violence among young girls is described. In this connection, the concept of Dialogic Feminist Gatherings is presented. Second, we describe the method we adopt: the Vignette Test. Third, the primary results of the data analysis are presented and discussed, in particular the significant decrease among the respondents in the wish to hook up with violent young men. The article concludes by presenting the insights that emerge as a result of the study and their relevance for future research on the prevention of gender violence among young women.

Attraction to Violence
In recent years, the literature that analyses the preferences of young women with respect to men and the factors that influence their relationship choices has increased. One study from the 1990s that has been frequently cited internationally was developed by Pyke (1996). Pyke presented a theoretical framework that addressed interpersonal power as interdependent with structures of gender and class and examined the ways in which interclass male dominance is exercised and reaffirmed. In a sense, Pyke analysed the question of power, as subsequent researchers have noted. Levi-Martin (2005) investigated the link between sexiness and power based on previous studies that illuminate this synergy. The author concluded that for both men and women the interpersonal power of men what makes them sexy. In a study that resonates with Levi-Martin's research, Rebellon and Manasse (2004) used data from the National Youth Survey (United States) that suggests that delinquency is linked to an increase in dating among perpetrators, with the possible effect of reinforcing delinquency. The perpetrator profile departs substantially from that presented in studies that examine the nice-guy stereotype (Urbaniak & Kilmann, 2003. Such studies offer evidence regarding the desirability of such men among college women for serious relationships but not for casual sexual relationships, for International and Multidisciplinary Journal of Social Sciences, 5(2) 187 which they choose "macho men" over "nice guys". In this area and at the time that Urbaniak and Kilmann published their studies, McDaniel (2005) developed a study on the factors that influence the young woman´s desire to date "nice guys" although preferring to date "jerks". The study carefully investigated the attractive or unattractive traits that the woman perceives a man to possess. In fact, the attraction element at the basis of McDaniel's study is at the core of other studies on young women, such as Bukowski, Sippola and Newcomb (2000). These authors focused on the transition from elementary to middle school and explained that the attraction to aggressive boys among girls increases during that period. However, less is known regarding how the attraction to violence may cause gender violence and how emerging insights might illuminate this synergy.
Since the early 2000s, the literature has identified models of attractiveness linked to violence against women. Gomez´s innovative study on the socialization processes in the sexual and affective relationships of adolescents (2004,2014) pioneered such research at the international level and influenced developments in this area (Padrós, 2012). Through the study of attraction linked to sexual-affective relationships, the identification of elements of change in contemporary love theories, the analysis of sexualaffective relationship models, and the inclusion of adolescent voices, Gomez argued that love is the result of a socialization process. Thus, although certain theories link love and attraction to instinct, an irrational emotion or "sexual chemistry" (Beck & Beck-Gernsheim, 1995;McDonald, 1998;Salecl, 1998;Sternberg, 1998), Gomez argued that any feeling is subject to change, which facilitates an alternative model of sexual-affective attraction and choice free of violence. Since the publication of his study, further research has provided additional evidence of this link and the ways in which it can be overcome to contribute to preventing gender violence among young adolescents.
Other studies that adopt this approach have investigated the link between attractiveness and violence. It is important to highlight the study by Valls, Puigvert and Duque (2008), which demonstrates that certain adolescents have been socialized into forming a type of relationship that results from this association. These authors performed their study in the Spanish context while drawing on previous research that reports a socialization process 188 Puigvert -Female Univesity Students Respond to Gender Violence within adolescent contexts that enhances this link. Additionally, their research presents evidence regarding how gender violence among young girls includes not only the partner or ex-partner but also the relations that occur during dates. These studies suggest that this link is a possible factor that influences the high rates of gender violence among young girls.

Prevention of Gender Violence against Young Girls
Research on the prevention of gender violence against adolescents includes a diversity of topics connected with this scourge. Several studies emphasize the importance of safe, healthy relationships. In this connection, several authors report that the period from 11 to 14 years is important time during which to enhance skills and attitudes that promote healthy relationships and reduce teen dating violence (Debnam, Howard & Garza, 2014). Other authors examine how to teach girls aged 15 to 18 years the characteristics of healthy relationships to reduce the risk of teen dating violence (Miller et al., 2015). Other studies focus on young men with the aim of reducing offensive behaviour and increasing awareness of dating violence among such men. For example, a follow-up on a prevention programme for high-school male athletes demonstrated that the perpetration of dating violence was less prevalent among the boys who participated in the intervention compared with control athletes (Miller et al., 2013). Additionally, other authors have studied the synergies between culture and violence in reports on dating violence in particular communities. Using data from the 2010 Minnesota Student Survey, Kast, Eisenberg and Sieving (2015) examined the role of parent communication and connectedness in dating violence victimization among Latino adolescents. The study concluded that parent communication and parent caring can play a role in preventing dating violence victimization among Latino youth. Inspired by these studies, several interventions and programmes have been developed to reduce the problem of gender violence that affects young girls or the risk of suffering such violence.
Intervention programmes for young girls to prevent gender violence have focused on various elements, such as the recognition of violence (Pick et al., 2010) and the awareness of services for victims (Foshee et al., 1998). In Canada, the Girls Action Foundation promotes a programme to create girls groups to discuss violence and other topics. These groups are based on an understanding of violence as a social problem, not an individual one, and include a feminist approach to social justice. These spaces for girls and with girls promote the acquisition of abilities to reduce violence not only in the lives of participants but also in the lives of other girls. In the United States, the initiative Love is Respect includes a collaboration with the Break the Cycle non-profit organization, which is dedicated to providing a comprehensive dating-abuse programme for young individuals aged 12 to 24 years. Among the organization's initiatives is a programme entitled "Host a party", which provides advice regarding how to discuss abuse and encourages the elimination of dating abuse by raising awareness in an informal environment. The effectiveness of such programmes has been addressed at the international level, and several efforts have been made to analyse their impact. For example, Lundgren and Amin (2015) reviewed interventions for girls and boys aged 10-19 years to prevent adolescent intimate partner violence and sexual violence from men against women in heterosexual relationships. To identify the need for additional evaluation, they classified programmes as effective, emerging, ineffective or unclear. In addition, Leen et al. (2013) reviewed the efficacy of intervention programmes in North America that address adolescent dating violence. In several cases, the authors identified significant long-term benefits, particularly in programmes focused on behavioural change. However, they note that a lack of programme repetition may undermine this benefit. Therefore, there is a concern regarding how to measure the impact of gender-violence prevention programmes among young girls and the consequences of such programmes for their lives.
This study contributes to this literature, including international research advances on the language of desire and its potential to contribute to preventing gender violence. The study utilizes research on interactions that socialize into desire and the socialization patterns behind such interactions. The development of an evidence-based intervention with young girls must consider the rates of gender violence in this group and investigate the causes underpinning this problem, raising awareness of the attraction to violence. Civil society, politicians and other social agents should seek effective methods to prevent the occurrence of this problem among young individuals 190 Puigvert -Female Univesity Students Respond to Gender Violence to decrease the high rates that develop later in life. Accordingly, this article provides evidence through a particular intervention.

Dialogic Feminist Gatherings
Dialogic Feminist Gatherings (hereinafter, DFGs) are egalitarian dialogues focused on the transformation of the language of desire to create possibilities for women who wish to question desires imposed by patriarchal societies by orienting these desires towards non-violent relationships (Flecha & Puigvert, 2010).
The gatherings are based on the following: 1) feminist contributions, particularly those oriented towards overcoming violent gender relationships; 2) scientifically validated contributions to overcoming violence; 3) scientifically validated analysis of hegemonic socialization of the desire towards violent men for sporadic or stable relationships; 4) scientifically validated contributions regarding programmes that succeed in preventive socialization; 5) egalitarian dialogue encouraged by persons who do not possess hierarchic status with respect to the women who participate in such gatherings.
Most interventions aimed at preventing gender violence resonate with the language of ethics, whereby adults, researchers or other professionals tell adolescents what is correct or good for their relationships and what is not. However, research based on the language of desire remains underexamined.
Therefore, this study is based on how in certain dialogues regarding young men who are considered non-violent the language of desire is missing and how when a conversation concerns those who exert domination over young women and practice violence the language of desire is present. The following statement, which was found in a youth magazine, defines well the separation between what a young woman's parents tell her to do (i.e., the language of ethics) and what the young woman actually does (i.e., the language of desire): My parents tell me to marry a good boy, and I really do what they say; before I marry, I'm having fun with the bad boys. This study seeks the transformation of the language of desire towards non-violent relationships.
International and Multidisciplinary Journal of Social Sciences, 5(2) 191

Method
The research followed a quasi-experimental design and was conducted at a public university in Spain and included heterosexual female students aged 18-27 years, who were mostly white and born in Spain (Table 1). This quasiexperimental study was based on the implementation of pre-and post-tests intended to evaluate the effect of a one-time intervention known as DFG on the preventive socialization of gender violence. The sample consisted of 127 subjects divided into three natural groups. These three natural groups were three classes with different professors: Group 1 (n=48), Group 2 (n=55) and Group 3 (n=24).

Instrument and Intervention
The instrument employed for the experiment was the Vignette Test, which is commonly used in studies that analyse behaviour and ideas regarding attractiveness and sexual attitudes in heterosexual adolescents (Ha et al., 2010;Santor, Messervey & Kusumakr, 2000). However, the Vignette Test designed in the framework of Free_Teen_Desire has a different, groundbreaking objective: to measure how women's desire changes through the implementation of DFG. The Vignette Tests are based on 4 pictures of young men that are accompanied by short texts that describe the men, particularly in terms of their attractiveness and their attitudes towards women. Two of the 4 pictures and descriptions portray young men who behave violently (Vignettes 1 and 3) and whose attitudes are considered in the literature to exemplify violence against women (Banyard et al., 2005;Fisher, Cullen & Turner, 1999;Gross et al., 2006;Kalof et al., 2001). In contrast, the other pictures (Vignettes 2 and 4) portray men who are not violent or sexist and whose attitudes are linked with kindness. However, during the experiment, in the post-test, we switched the pictures of Man 2 and Man 4 from non-violent to violent profiles. In the end, a total of 12 vignette profiles (including violent and non-violent profiles) were designed. All of the vignettes included four questions regarding whether the participants and their female friends would be interested in forming a relationship or hook up with the young men portrayed in the vignettes. The respondents answered these questions using a Likert scale (1 to 3 signified 'No'; 4 to 6 signified 'Yes'). This article presents the results for two of the questions: "Would your friends like to hook up with him in a party?" and "Would you like to hook up with him in party?". As noted, the one-time intervention employed was the DFG. In the experiment presented here, the knowledge used in the DFG is based on data collected in the research project Mirage of upward mobility (Oliver, 2010(Oliver, -2012

Procedure
The implementation of the pre-and post-tests followed three steps regardless of group. There was a common pattern as follows: International and Multidisciplinary Journal of Social Sciences, 5(2) 193 a) Students complete the pre-test with two vignettes that depict young men with violent behaviour opposite two vignettes that depict young men who are not violent. The average time required to complete the pre-test was approximately 10-15 minutes.
b) The DFG is performed. The average time spent on this intervention was approximately 30-45 minutes.
c) Students complete the post-test with different vignettes, in which four young men are described: two violent and two non-violent. The average time required to complete the post-test was approximately 10-15 minutes.
The procedure for Group 1 was to complete the pre-and post-tests the same day but the post-test after the students participated in the DFG. Group 2 completed the post-test one week after completing the pre-test and participating in the DFG. Finally, Group 3 participated in the DFG and completed the post-test two weeks after completing the pre-test (after a twoweek break from classes).

Data Analysis Strategy
A descriptive analysis was performed on the participant responses to the preand post-test questions. The responses of male students were eliminated from the data matrix. However, a comparison of column means was performed (i.e., a T-Test) ( Table 2) to determine if a change in tendency regarding attraction towards violent young men in the pre-and post-tests occurred.

Results
Based on the previously described quasi-experimental design, the results that we present are primarily based on descriptive frequencies that compare the desire towards violent men when the DFG is implemented. According to the T-Test results, we can reject the null hypothesis, which signifies that the means of the pre-tests and the post-tests differ. That is, the DFG made a difference in the women's answers between the preand post-test. This outcome is true for all cases except Pairs 4 and 6.
Based on these results, we confirm that there are differences between the pre-test and the post-test. This outcome suggests that the DFG affected the women's answers. Finally, we would like to state that a statistical grouping was realized with the Likert scale to obtain a dichotomous variable.
Next, we present the results for each of the classes on whom the experiment was performed. Table 2. T-Test results for Groups 1,2,and 3 International and Multidisciplinary Journal of Social Sciences,5(2) 195

Group 1
According to the pre-test data for young women (n=48), the responses of these participants changed importantly when the one-time intervention (i.e., the DFG) was performed. When these participants were asked about their female friends (pre-test), 44 respondents stated that they believed that their friends were interested to hook up with a violent male. Later, after the DFG was performed, the number of positive answers related to the vignettes that depicted violent males substantially decreased to 23 responses (post-test). Thus, the DFG directly affected the women's views regarding the desire of their friends with respect to this male type. In fact, a difference of 52.2% occurred.
When these results are compared with how women directly answer regarding their own desire towards violent young men, the pre-and post-test data reveal a relevant change. In the pre-test, 16 women (22.5%) affirmed that they would hook up with a violent young man. However, in the posttest, only 6 respondents (8.4%) stated that they would hook up with a violent young man. In this case, the difference was 37.5%.

Group 2
The procedure for this group resembled that for the previous group except that the DFG was implemented one week later. However, for Group 2 (n=55), a similar tendency was confirmed. Therefore, time is not a variable that interferes with the change of desire. Thus, there is a consistent pattern. In this case, there were 49 responses to the question whether one's female friends would desire to hook up with a violent young man (45.7%).
However, after the DFG was implemented (n=45) 3 , the number of responses substantially decreased to 30 responses (56.6%). This transformation implies a difference of 61.2%.
In contrast, regarding the women's own attitudes towards violent males, 25 respondents stated that they would hook up referring with a violent young man. However, after the one-time intervention, this number decreased to 13. In this case, the difference was 52%.

Group 3
This last group completed the post-test two weeks after the pre-test following a break from classes, which occurred in the middle of the experiment (n=24). For this group, a difference appears that is not relevant with respect to our overall results but that we want to highlight. According to the pre-test data, 7 respondents stated that their friends would be interested in hooking up with violent young men. However, on the post-test, this number increases to 9. To explain this difference, additional research is required on what occurs during holidays and whether this phenomenon affected the small increase in positive responses in contrast to the decrease exhibited by the other groups. In addition, a relevant change could be confirmed: when the surveyed women answered regarding their own interest, in the pre-test, 8 students affirmed that they would hook up with a violent young man. In the post-test, none of the students wanted to hook up with this type of man.
In sum, the groups demonstrate the effect of the DFG in changing the attitude of young women towards violent young men (Table 3). In fact, overall, there is an important change in percentages when the women discussed their female friends (78.4%) and themselves (48.8%) in the pretest compared with in the post-test (38.5% and 14.9%, respectively). Under the framework of the Marie Curie Action, additional analysis and data are in preparation. International and Multidisciplinary Journal of Social Sciences, 5(2) 197

Discussion
A highly significant result of the analysis is the substantial impact of the DFG on the desire of young heterosexual women. As it can be observed in the data collected on the three groups (see summary in Table 3), when this intervention was implemented, the number of positive answers regarding attraction towards violent profiles substantially decreased. These results reveal the social dimension of attraction and desire and therefore the impact of interventions in questioning attraction to violence and preventing gender violence. The results indicate change in desire with respect to the attraction of young women to violent young men. As noted in the literature, the language of desire plays a central role in this transformation. However, to demonstrate how long the effect endures and to identify the processes that can increase or decrease the effect over time, additional research is required.

Conclusion
As described in the literature review, research on gender violence has revealed the close connection between violence and attractiveness (Pyke, 1996;Rebellon & Manase, 2004;Urbaniak & Kilmann, 2003McDaniel, 2005). This phenomenon significantly contributes to understanding the high percentages of harassment and sexual violence against women worldwide (UNICEF, 2014;FRA, 2014). The literature has also identified strategies that contribute to preventing this problem among adolescents, such as parental communication with girls, awareness of victim services and spaces for dialogue addressed to young women aimed to help them acquire abilities to prevent violence (Kast, Eisenberg & Sieving, 2015;Pick et al., 2010;Foshee et al., 1998). However, to provide mechanisms to eliminate the attractiveness of violent males, additional research is required.
The results presented in this article will contribute to these discussions and the reconsideration of the scientific knowledge employed in the design of interventions to prevent gender violence. This article provides new insights regarding the relevance of interventions and policies that are based on the transformation of the language of desire. As can be observed in our