Attitudes of High School Students Regarding Intimate Relationships and Gender Norms in New Providence, The Bahamas

This paper reports the attitudes and actions on relationships with the opposite sex of 1,002 Grade 10 and Grade 12 students in New Providence. Girls were more likely than boys to use aggressive behaviours in teen relationships. Some of the behaviours noted in teen relationships informed expectations of marital relationships, such as restricted access to friends of the opposite sex. The students endorsed a number of sex-related stereotypes, such as a man being the head of the household. Both male and female students indicated that it was acceptable for men to control their wives. Participation in aggressive and controlling behaviours by teens points to the need to educate students about how to develop more respectful relationships


INTRODUCTION
In common with countries in the Caribbean and elsewhere, The Bahamas has been subjected to increased levels of violence, has a relatively high incidence of rape in the of concern, and sexual relations can be risky due to unprotected sexual intercourse (Yu et al, 2007).
Teen relationships can result in live births; 617 in 2008, 532 in 2009and 507 in 2010(Bahamas Department of Statistics, 2012. 2 While the number of reported rape cases has been generally declining since 2006, the number of cases of unlawful sexual intercourse has not, as illustrated in Figure 1. In The Bahamas, anecdotal information from guidance counsellors, social workers, the Royal Bahamas Police Force and the Bahamas Crisis Centre suggests that increasing numbers of children are experiencing sexual abuse and/or abusive intimate relationships (Concerning Child Abuse Figures, 2014). In 2011, 167 child sexual abuse cases and 11 incest cases were reported (Bahamas Department of Social Services, 2012). Between 2007 and 2011, the number of clients seen at the Bahamas Crisis Centre moved from 770 to 1,285. Over the same period, the number of new cases of children (those under 18 years) seen in the Bahamas Crisis Centre declined slightly from 96 in 2007 to 81 in 2011, but the percentage of those who were males increased, from 38% in 2007 to 46% in 2011. The longer-term effect of sexual abuse of teens can result in their having attitudes about intimate relationships that permit abuse, including domestic violence (Barter, McCarry, Berridge & Evans, 2009).
In the United States, teenagers experience violence in their intimate or dating relationships. The prevalence of physical dating violence was around 9% from 1999to 2003, and 10% from 2004(Henry & Zeytinoglu, 2012. In the Caribbean, similar findings have emerged. Le Franc et al. (2008) claimed that in Barbados, Jamaica and Trinidad and Tobago, that "a culture of violence and adversarial intimate relationships may be well entrenched" (p. 409).
The UNICEF State of The World's Children 2011 report states that physical, sexual or psychological acts of violence take place within the home, at school, and in the community. The full scale of violence against adolescents is impossible to measure given that most abuse occurs in secret. In every country assessed, violence was an important problem. Additionally, it was noted that more than one third of adolescent males aged 15-19 considered a husband justified in hitting or beating his wife. In The Bahamas, 11% of college students reported that it could be acceptable for a husband to slap his wife while only 4% of students from the United States of America felt this way (Risley-Curtiss & Fielding, 2012). Excessive controlling behaviours put females at risk of femicide and demonstrate the ultimate outcome of inadequately formed relationships (Fukuroda, 2005).
These observations make it important to ascertain the attitudes of teens regarding relationships, particularly as many are exposed to violence at home (Carroll, Fielding, Brennen & Hutcheson, 2011) and at school (Reynolds, 2009) although a relationship between parental values and those of children in The Bahamas is yet to be demonstrated (Lui et al., 2007).

Purpose of study
This study was designed to investigate the perspectives of teenagers in The Bahamas on interpersonal relationships and the prevalence of violence experienced in relationships, particularly those of a romantic nature. Relationships considered intimate can take many forms (Coy & Miller, 2014), but for the purpose of this study we consider these relationships to be emotional/romantic in nature and not necessarily sexual. Tharp et al. (2011) remind us that teen dating violence is a preventable public health problem and that dating violence typically starts at school, so making interventions at this stage is important. Appropriate training during school years can promote healthy relationships with their attendant long-term benefits (Crooks, Chiodo, Zwarych, Hughes, & Wolfe, 2013).

METHODOLOGY
The target population for this study was 1,000 adolescent boys and girls in Grades 10 and 12, thus covering those aged about 15 to 18 years. Students in Grades 10 and 12 were chosen for this study as these were the grades in which the Bahamas Crisis Centre was making its presentations on healthy teen relationships. Only those students who participated in the training given by the Bahamas Crisis Centre were surveyed. All government high schools and selected private high schools in New Providence were included in the study.
A team at the Bahamas Crisis Centre developed the questionnaire with assistance from the Health Information Unit of The Bahamas Ministry of Health.
The Liz Claibourne on-line forum (Teenage Research Unlimited, 2006) and training facility on teen relationship violence guided the development of some of the questions. Other questions were taken from Weller and Clifford (2008). The questionnaire collected information on demographics and information on relationships. Volunteers were trained to present a healthy relationship script prepared by the Bahamas Crisis Centre and the Zonta Club. The questionnaire was administered before the presentation to the classes of Grades 10 and 12 students. The survey took about 10 minutes to complete. The process was monitored by the presenter and, in some cases, a teacher. Permission for this study was given by the Ministry of Education, Youth, Sports and Culture. Interviews took place during 2011.

RESULTS
The total number of respondents was 1,002 from eight schools, which included one private school. Not all forms were completed, so this represents the maximum sample size. A minority of students were male, 45.2% of 960 respondents, and 13.1% of 945 students were in Grade 10. The mean age of the Grade 10 students was 15 years, SE = 0.74, and that for the Grade 12 students M = 16.8 years, SE = 0.03.

Family structure
The majority, 88.1%, of students lived in homes with a mother figure (mother or stepmother) and over half lived in homes with a father figure, 53.2%, as illustrated in Table 1.
Students who lived with both father and mother figures were in the minority, 40%; 21.7% lived with a grandmother. When a grandmother was in the home, a father figure was less likely to be present than a mother figure as illustrated in Figure 2. We use the term father/mother figure to include either a biological or non-biological parent as in Coffino (2009); no formal legal status of the person to the child is assumed: for example, a father's girl-friend or step-mother may take the place of the absent biological mother.  Boys typically did not date girls older than themselves, whereas the opposite was the case with girls. Only 7.1% of boys dated a girl from an age group older than their own, compared to 30.5% of girls (Table 3).
Relatively few girls dated boys who were much older than they were. Boys were more likely than girls to think that it was good for their girl/boyfriend to be controlling and both boys and girls were equally likely to think that it was acceptable for their girl/boyfriend to be jealous and want to call often (Table 4).  Both boys and girls were subjected to controlling behaviours by their partners, as shown in Table 5, but not always to the same extent. Girls were generally more aggressive than boys in their relationships, but both were equally unlikely to use weapons as a means of coercion. See Table 6 below for details. Using the attitudes in Table 5, a logistic regression investigated the association with actions in Table 4, which occurred more than once in experienced relationships, and the sex of the respondent. A stepwise backward elimination regression (using p > .10 level for elimination) was used. Only acting jealously and thinking that it is good for a partner to be controlling were related to the variables listed in Table 6, as reported in Table 7. Although sex was not linked to attitudes towards jealously, slapping was linked. The use of weapons was rare (as seen in Table 6) but their use was linked with attitudes towards controlling behaviours (Table 7).

Attitudes towards future relationships/actions
There were consistent differences between boys and girls in their attitudes to adult relationships but these attitudes were also endorsed by both girls and boys. Despite the differences in the responses between the sexes, the majority of boys and girls subscribed to traditional ideas that men should be the heads of their households and control their wives. The expectation that a wife should have sex with her husband whenever he wants was correlated with the expectation that the man must be the head of the household, for both female (Kendal's τ(494) .128, p < .001) and male respondents (Kendal's τ(398) .113, p < .001). Table 9 illustrated the need for mutual respect by both partners was indicated by the similar responses for the need for both parties to submit to each other. Both boys and girls expected their wives/husbands to stay with them (i.e. remain married) even if the relationship was not going well. Although the percentages were not large, boys and girls accepted that men could be violent towards their wives as a form of expressing their love and that women must have sexual intercourse with their husbands on demand. Aspects of their current relationships, such as jealousy if time was spent with others and the expectation that other friendships would be limited (Table 5), were expected to continue in their adult relationships.

Link between violence in experienced relationships and expectations for adult relationships
For those attitudes in Table 8 which were violence-related, a logistic regression was performed to determine if agreement expressed by respondents was related to actions in Table 6, which occurred more than once in experienced relationships. A stepwise backward elimination regression (using p > .10 level for elimination) was used. The analysis indicated that after the respondent's sex was associated with their attitudes, the calling of names was the second most important factor. Use of a weapon was only associated with "slapping up" a woman "to show who is boss".  Today, violence in homes challenges the longstanding belief that homes and family are places and sources of love and protection (Carroll, et al, 2011).
In this study, typically both parental figures worked. This observation blurs the traditional view of men being the breadwinner, or at least the sole breadwinner. Further, when male parental figures were absent from the household, they tended to be replaced by a grandmother, resulting in respondents being more likely to be brought up in a female, rather than male, dominated environment, and so be exposed to traditional views. This is consistent with the matrifocal nature of Bahamian homes (Bethel, Minnis, & Fielding, 2012). Children in female-headed households are at greater risk of living in poverty (Bahamas Department of Statistics, 2004Statistics, & 2014, so this may account for the fact that boys are encouraged to start work to contribute to household income. One important sex difference, which may have influences beyond teen years, was that boys were more likely than girls to be engaged in employment. Consequently, from an early age, gender conditioning inculcates the view that the man will provide for the household and with that economic responsibility, certain rights are also earned (World Bank, 2012). The early introduction of boys to the market place, and its associated economic implications, may explain why boys expect themselves to be heads of households, rather than girls. However, it should be recognised that the need to earn money can put children at risk of being abused, for example through sexual exploitation (Dunn, 2002).
Over 80% of respondents had been on a date and so had had a relationship of some sort with the opposite sex; this figure is higher than the 61% reported in the United States of America (Teenage Research Unlimited, 2006). While this study cannot answer why this may be, it could be that if students do not receive support from homes or schools they may look to others for comfort and support.
While not the focus of this study, we acknowledge the importance of adult support to children. In the Bahamian context, this represents an area which requires further research as previous studies have only obliquely visited this matter (Bahamas Ministry of Health, 2003 andMinnis, Symonette, Newry, Gibson &Stevenson, 2011). It also suggests a lack of adult support for children.
Relatively few girls dated boys outside of their age group. Although this aspect of teenage behaviour has been reported as an area of concern in the news media (for example  April 2), it appears that girls dating boys slightly older than themselves was the norm. Barter et al. (2009) noted that when girls date boys two or more years older than themselves, they are at risk for becoming victims of emotional violence. Future research on those girls who do date older males and the circumstances surrounding such relationships is recommended.
A minority of teens approved of their partners being controlling, but contradicted this by their attitudes towards behaviours that could be considered as controlling. This may suggest that teenagers do not have a full appreciation of what behaviours are and are not controlling.
The percentages of respondents in our study subjected to controlling behaviours were higher than those reported by teenagers in the United States (Teenage Research Unlimited, 2006). McEwen (2010) considered Bahamian homes to be subjected to "high levels of violence" (p. 12) compared to the United Kingdom, and this violence may influence teen interactions.
In the United Kingdom, Barter et al. (2009) found that girls were more likely to exhibit controlling behaviours than boys, a finding similar to that of our study, but this sexrelated behaviour was less pronounced than in the United States (Teenage Research Unlimited, 2006) and contrary to a previous study in The Bahamas (Bahamas Ministry of Health, 2003) 3 . The data do not support the idea that boys are necessarily the leaders in teen relationships. If this is so, it presents a reversal of actions of adults where males are reported as controlling relationships and even fighting over women (Smith, 2012 June 5).
The controlling behaviours exhibited by teens can be expected to persist into adulthood. Students were concerned about (future) spouses having friends of the opposite sex. They indicate that women should do as they are told, or at least conform to the wishes of their male partners (e.g. dressing appropriately). The conservative culture of The Bahamas, often underpinned by the influential Bahamas Christian Council, a coalition of more recently established Christian sects (Bahamas Christian Council, n.d.), reinforces norms which include male leadership in homes.
Therefore, these attitudes may moderate the attitudes regarding leadership in the household of women who are married or in long-term relationships. Attitudes concerning controlling behaviours include the expectation of both boys and girls that wives must always be willing to have sex with their husbands. Although there were sex differences in the responses, 16% of the girls thought that wives should provide their husbands with sex on demand. This attitude has implications for the debate in The Bahamas concerning a marital rape law (Benjamin & LeGrand, 2012). Our data suggest that sexual expectations of future husbands and wives are formed at an early age, and this raises the question as to the origin of these expectations.
While married couples were expected to remain together, this contrasts with the reality within the respondents' homes as only 40% of respondents lived with both father and mother figures. Behaviours associated with men and women were often based on stereotypes. For example, women were expected to stay at home, but for most respondents the mother figure worked. Consequently there is a disconnection between reality (students having mothers who worked) and attitudes ("a woman's place is in the home"). These expectations may reflect the conservative nature of Bahaman society which is underpinned by fundamentalist Christian beliefs and orthodox statements by public figures (Nicolls, 2012).
The use of violence by female teenagers appears to be consistent with the negative image sometimes projected onto women through the Jezebel factor which portrays women as "sexually aggressive, immoral [and] domineering" (Marshall, 2006, p. 2). In The Bahamian context, the term "junglist" is applied to "very vulgar, very loud, very irreverent" women (Turnquest, 2012) who also use their sexuality for their own purposes. These images can provide an excuse for men to feel that they need to control "their" woman, as demonstrated by the perceived need by males to hit their partner to show who is in charge.
Attitudes towards violent behaviours in adulthood were linked only to a limited extent to violent behaviours in current relationship. The fact that "calling names" was linked with attitudes in adult relationships raises questions concerning what this action entails. Children are subject to being shouted at when at home (Carroll, et al, 2011) so it is something with which many children will be familiar. This suggests that name-calling within a relationship holds particular significance beyond that experienced at home. In the United States, name-calling has been shown to have important detrimental effects on teenagers (Teicher, 2010) and the importance of the physiological aspects of defective teen relationships has been demonstrated by Nansel et al. (2001).
The responses show that on a number of issues regarding relationships, boys and girls have different attitudes and behaviours. It can also been seen that large numbers of teens can be expected to be victims of controlling behaviours. The use of threats and physical force may be learnt behaviours due to the presence of violence in homes in The Bahamas. Children are sometimes subjected to violence from an early age and so they have ample opportunity to learn violence both at home and at school (Carroll et al., 2011).
The ultimate form of violence is murder. With arguments and domestic violence being key reasons for murder in The Bahamas (Hanna, 2010), these data suggest that there is a clear need for children to be taught how to respect one another from an early age. Again, the higher percentage of respondents in this study, compared to the percentage in a study in the United States (Teenage Research Unlimited, 2006), who threatened to use murder to maintain a relationship might suggest either that the relationship is deeply felt, or that students are anxious to maintain the relationship, no matter what. This may mean that they will have to unlearn behaviours learned from their peers. The treatment of pupils at school by teachers can either reinforce that violence is an effective form of control (McNeil, 2012) or provide teaching moments in conflict resolution. While teaching healthy relationships in schools can reach the majority of teenagers, over 1,000 teenagers do not attend school (UNESCO Institute for Statistics, 2013) and so are beyond the reach of instruction.
Overall, it is apparent that the breakdown in adult relationships, which is considered to play an important role in the violence observed in Bahamian society, may be the consequence of adolescents not adjusting their teen behaviours when they become adults. Therefore, modifying the attitudes of children with regard to interpersonal relationships may be important in reducing long-term violence in the country. Boys and girls had different attitudes on many aspects of relationships with current or future partners, but their endorsement of stereotypes of sex-related roles and their participation in certain behaviours could be a cause for concern. Underpinning these attitudes may be issues associated with what it means to be a woman or a man in The Bahamas (Strachan, 2013), and related gender norms. The fact that students found violence of any type acceptable in relationships may indicate their "normality" in Bahamian society.
The differences in responses exhibited by boys and girls may need to be taken into account when teaching students about how to manage relationships. Sessions for the two sexes may need to be held separately, as well as jointly, depending on the material to be covered. Teaching teenagers about healthy relationships allows for differences in opinions (for example Lilley, 2003) but sexbased differences may pose additional difficulties in accommodating such sessions in the co-educational system that prevails in The Bahamas.

CONCLUSIONS
This study has informed the dialogue towards the appreciation of the nature of relationships between teenagers in The Bahamas. Both boys and girls subject each other, and are subjected to, controlling behaviours and violence in their relationships, with girls being no less aggressive than boys. The study indicates that teenagers expect males to be the controlling partners in adult relationships, and that women should conform to the stereotype of a housewife. Interventions at school can be useful in attempting to alter teenage relationships so that peer-on-peer abuse is reduced. However, culturally sensitive training can only be implemented once there is an understanding of the nature of these relationships, and of the expectations of teenagers for their adult relationships. Further research is required to determine the sources of teenage attitudes towards relationships and the impact of gender norms. Unless these are known, interventions at school may be both undermined and ineffectual.