Pilot.

The questionnaire was piloted in a school with a similar setting, but not included in the study, amongst 42 young women (nearly 10%), to ensure clarity of the questions and consistency in the methods of questioning and the data collection procedure. After the pilot, some issues relating to the demographic information were re-formulated for the school-going population in an urban setting in Maputo.

Instruments.

Prior to the survey an exploratory study using focus group discussions was conducted and themes were generated to explain the IPV experienced by young women in Maputo city. The questionnaire for the survey was adapted informed by the themes and based on the WHO Multi-country surveys of violence against women instruments (Garcia-Moreno, 2005). We also based our questions on the socioecological model (Bronfenbrenner, 1998) [30]. The themes that emerged from the focus group discussions were integrated into the model for variables and included individual factors (Age of women, age of partner, young women’s relationship status, young women’s employment status; HOH education level, HOH employment status, partner employment status), community factors (religiosity) and societal factors (perceptions of gender roles).

The questionnaire used in this study to estimate the IPV and associated factors was adapted from the WHO Multi-country Survey of Women’s Health and Domestic Violence against Women (Garcia-Moreno, 2005). The WHO Multi-country tools are recommended since they cover issues of IPV and their validity and reliability have been confirmed [31]. The questionnaire’s validity in the Portuguese language was confirmed in the study done in Brazil in two different social contexts (urban and rural). The results indicated the adequacy of the instrument in estimating the occurrence of IPV and the associated factors. The study reported a Cronbach alpha coefficient of 0.88. Thus, the instrument has been shown to be reliable, consistent and adequate to be used in other similar studies accessing IPV, in different contexts such as this study [32].

The questionnaire was translated from English (S1 Appendix) to Portuguese (S2 Appendix) and back translated into English by a second translator to ensure consistency. The selection of the questions was designed to address the sociocultural context of the young women attending secondary schools in Maputo, based on the information obtained from the focus groups and the available literature.

Our study was done in an urban setting and the instrument enabled us to estimate the different types of violence (physical, sexual and psychological). The IPV was measured both across the lifetime and in the 12 months prior to conduct the survey.

Dependent variables. The dependent variables comprised acts of physical, sexual and psychological violence, and the questions were adapted from the WHO Multi-country Survey of Women’s Health and Domestic Violence against Women (Garcia-Moreno, 2005), using the subscales of the Abusive Behavior Inventory of physical violence [33], psychological violence [34] and sexual violence [35]. Physical, sexual or psychological violence was assessed by questioning if, since the age of 15 the young woman had ever experienced one of the acts from a current or past partner. Experience of physical, sexual or psychological violence was considered confirmed if the response was “yes” to at least one of the defined criteria questions.

Physical violence (8 questions). For example: “Has he or any other partner ever slapped you or thrown something at you which could hurt you?”

Sexual violence (10 questions). For example: “Has he or any other partner ever physically forced you to have sexual intercourse?”

Psychological abuse (13 questions). For example:” Has he or any other partner called you insulting names?”.

The questions presented to participants to assess the occurrence of each type of violence are detailed in the questionnaire (S1 Appendix).

The lifetime prevalence of IPV was defined as the proportion of women who have or ever had an intimate partner and reported violence from a partner at any time in their life since the age of 15 years.

The current prevalence or 12 months prevalence of IPV was defined as the proportion of women who currently have or ever had an intimate partner in the previous 12 months before the survey and reported violence in the previous 12 months.

An intimate partner was defined as any male partner with whom the young women have or ever had a romantic relationship that included sexual activities, either spouse/husband, boyfriend/dating partner, or ongoing sexual partner/occasional partner. The definition was based on the sociocultural context of an urban setting in Maputo. After accessing the meanings young women attribute to intimate relationships, during the pilot. Therefore, we contextualized the definition to address the specific group during the survey.

Independent variables. This study is part of a larger study undertaken to assess the prevalence and risk factors for IPV among young women in Maputo city. Prior to the survey an exploratory study using focus group discussions was conducted and themes were generated to explain the IPV experienced by young women in Maputo city. Therefore, the questionnaire, based on the WHO Multi-country surveys of violence against women was informed by the contextual themes and the literature concerning IPV against women. For example, the focus group discussed themes of religiosity and IPV. Therefore, we included the religiosity as a contextual variable to investigate how religion might shape young women’s perspectives, beliefs and the influence on IPV. The variables were also informed by the available literature to explain the risk factors for IPV in different sociocultural contexts. In settings such as Mozambique the society is dictated to by social norms which give privilege to male dominance [21, 23, 36]. Thus, since contextual sociocultural factors may explain the occurrence of IPV, we included sociocultural variables. We also included questions from the socioecological model which established that the risk factors for IPV emerge from different constructions within people’s interactions. These included individual, relationship, community and societal factors (Bronfenbrenner, 1998). The independent variables were divided into two sections. Section one comprised socio-demographic characteristics measured as categorical variables and section two investigated the sociocultural risk factors for IPV considering agreement or disagreement with statements of male superiority and the statements of acceptance of IPV. These were measured on a 4-point Likert scale from strongly agree, agree, disagree and strongly disagree [37].

Section one of the questionnaire comprised questions about:

Socio-demographic factors: Age, divided into two categories (15−19, 20−24 years); Employment status of respondent; Relationship status of respondent; Commitment to religion: Defined as the degree to which a participant adheres to his or her religious values, beliefs, and practices and uses them in daily living. These were measured as yes/no responses.

Household factors: With whom the participant lives; by whom the participant was raised; Educational level of the Head of Household (HOH); Employment status of the HOH.

Partner background: Age differences between the young women and partner; Alcohol consumption by partner; Employment status of the partner.

Section two comprised sociocultural variables measuring risk factors, and examples of the questions are provided:

Perceptions of gender roles (8 questions): For example: “Do you believe that a man has a superior position within a society than women?”.

Tolerance of violence (8 questions). For example: “There are times when violence by men to women is okay”.

Lifetime prevalence of IPV.

Of the 413 young women who provided information about their experiences of IPV, 248 (60%) (95% CI: 55.15–64.61) had experienced at least one form of IPV in their lifetime. More than half of the young women, 230 (55.7%), had experienced at least one act of psychological violence, 120 (29.1%) had experienced at least one act of sexual violence, and 93 (22.5%) had experienced at least one act of physical violence. Of the study participants 45 (10.9%) had experienced all three forms of violence, with co-occurrence of two forms of violence reported by 55 (13.3%) for psychological and sexual violence, 32 (7.4%) for psychological and physical violence and 5 (1.2%) for sexual and physical violence. There were 361 women in the 15−19 years age category and 209 of them had experienced IPV constituting a proportion of 57.9%, whereas of the 52 women in the 20−24 age group, 39 of them experienced IPV constituting a proportion of 75.0% of that age group. In general, the overall IPV increased with increasing age, with psychological violence predominating in both age categories.

12 months prevalence of IPV.

Of the 293 young women who had a partner in the previous 12 months, 186 (63.4%) (95% CI: 57.68–69.00) reported IPV in the 12 months prior to the data collection, with psychological violence predominant, reported by 164 (55.9%) respondents (Table 2). Physical violence was reported by 55 (18.7%) respondents, and the prevalence of sexual violence was higher at 71 (24.2%).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Lifetime and previous 12 months prevalence of physical, sexual and psychological violence by age categories.

https://doi.org/10.1371/journal.pone.0243304.t002

As expected, the overall 12 months prevalence of IPV decreased with increasing age. In general, the younger women had a higher prevalence of current IPV as compared with the older age category. There were 238 women in the 15−19 years age category and 155 (65%) of them had experienced IPV whereas of the 55 women in the 20−24 age group, 31 (56.4%) of them experienced IPV. However, some variations in the pattern suggest that although the older women are more protected from psychological violence, they are likely to be more at risk of physical and sexual violence as compared to younger women.

The lifetime and previous 12 months prevalence of IPV are presented in Table 2.

Individual factors.

Age of young women. Our study revealed that in the bivariate analysis, the odds of being abused were lower among younger women (15–19), (OR, 0.458, 95% CI: 0.237−0.888, p = 0.021), as compared to those older (20–24) years. These findings were consistent with previous studies which reported increasing odds of IPV with increasing age among groups of young women in studies from SSA and elsewhere, and in the recent study conducted in Zimbabwe to assess the trends in IPV prevalence among ever-partnered women [21, 41, 48].

These results suggest that women in their early twenties may be more time exposed to relationship conflicts since they may be married or in committed relationships. They are perhaps more likely to face relationship conflicts and therefore to tolerate IPV and to remain in a relationship with violence, while the younger women are probably at the initiation stage of their relationships, and are therefore less time exposed to relationships and conflicting relationships, and may therefore be less likely to be committed to staying in such a relationship when there are conflicts.

Our multivariable analysis did not confirm the results of the bivariate analysis, but the high prevalence of IPV that was reported by the young women emphasizes the need for specific intervention strategies to address potential risk factors and to prevent the occurrence of IPV early in young women’s lives. For example, in Mozambique, a previous study among women aged 15–49 years attending schools and universities, revealed that adolescent and young women were more vulnerable to violence compared to women above the age of 24. The study’s findings also reported a positive association between their young age, their single status, their habit of going out to parties and their alcohol consumption, with increased risk of IPV [22]. In the context of South African and USA settings for example, the findings suggested that the young age of women reduced their ability to deal with the complexities and the dynamics of relationships, therefore increasing their risk for IPV [43, 45–47, 51]. In a study which explored the trends in prevalence and risk factors associated with IPV among Zimbabwean women of reproductive age (15–49 years), the older women above 40 years, were less likely to experience sexual and physical IPV compared to the younger women between 15–24 years [48]. This suggests that although the older women are more time exposed to relationships and may have experienced violence in the initial stages of their relationships, their experience regarding relationships may increase their ability to manage a more stable relationship and also to use more protective methods when in a conflicting relationship.

The likely reasons for the differences in the different studies’ results concerning the age of the women experiencing IPV, may be due to the selection process for the study samples and the comparisons between different age categories. Our study recruited only school-going women between the ages of 15 and 24 years. In Mozambique, a developing country, children often enter school when they are older than is typical in other countries and continue to attend school above the age of 20. Therefore, amongst these young women attending school to improve their educational attainment, their young age and their single status suggest that they may be less exposed to conflicting relationships, or that they can challenge and not tolerate conflict in relationships.

Further, the older age category in the study is less represented in the sample size and in this setting and this may have affected the results. Interpreting these results calls for caution and the consideration of the local context when developing and implementing IPV programs. The findings of this study suggest that the consideration of age as a factor associated with vulnerability to IPV, needs to be considered as a specific or contextual factor with potential variability across countries and settings.

These study results indicate that IPV is of concern in this urban setting in Maputo and emphasize the importance of early interventions in Mozambique, so that young women are made aware of the risk of IPV and are advised about how this could be prevented, and if violence occurs, how this should be handled. Efforts are required in order to enhance the young women’s reproductive rights, so that young women in schools and communities can pursue healthy relationships free from violence.

Young women’s HOH’s unemployment status, partner employment status and partner more than ten years older. In the bivariate analysis, several of the risk factors which have been reported from other studies showed a similar trend in this study. Young women whose partners were more than ten years older than themselves and those who did not know the age of their partner were more likely to report IPV (OR, 4.528, 95% CI: 0.986–20.805, p = 0.052).

Similar findings are reported from a Botswana study, which revealed that large age differences between partners is a predictor for IPV [52]. Moreover, studies from South Africa and other SSA settings reported that women who have partners older than themselves struggle to air their opinions about relationships, and furthermore, the older partner may expose young women to risky behavior that includes violence [41, 46, 51].

Economic status has also been shown to be an essential factor associated with IPV. In the multivariable analysis, there was significant association between the financial situation of the HOH, if unemployed, and young women reporting IPV (AOR, 1.642 95% CI: 1.044–2.584, p = 0.032). This suggests that young women from households with unemployed HOHs have an increased likelihood of experiencing financial constraints, and therefore dating older men may result from the economic dependence of women.

Moreover, the bivariate analysis suggests that having a partner who is employed (OR, 2.247 95% CI: 1.87−4.256, p = 0.013) might lead them to tolerate and to remain in relationships with IPV, since employed partners are more likely to exert financial power over the women, and may even perpetrate violence against them. This was also a confirmed trend in the multivariable analysis which showed that the risk of abuse increased when the partner is employed (AOR, 1.675 95% CI: 1.044–2.584. p = 0.155).

Our findings follow previous studies which reported that adolescent, single young adults experiencing financial strain were more vulnerable to being abused [22]. This is also consistent with findings from South African adolescent learners that revealed a high risk of IPV among those adolescents not receiving pocket money from their parents, compared to those who were [46]. Findings on low socio-economic status and the increasing risk of IPV were also reported in studies from SSA and elsewhere [41] and in a Zimbabwean and Ghanaian Studies [48, 53].

Similarly, a survey conducted in 31 countries, including South Africa and Tanzania, evaluating IPV and economic status among college students, revealed that the higher levels of IPV were associated with an inability to meet daily financial needs and with being younger [54]. In a study conducted in low-income and middle-income countries, the findings indicated that in general, richer and more empowered women reported less IPV [20]. A study conducted in South Africa found that IPV reduced with interventions combining education and economic empowerments. The violence experienced by women with their own income decreased by 55% (A0R = 0.45; 95% CI = 0.23, 0.91). There were also improvements in the women’s ability to challenge the acceptance of violence and to leave violent relationship [55]. The results suggest that improving the socio-economic circumstances of women may improve women’s independence and control over their lives, improving their skills to manage healthy relationships with no violence. Particularly among younger women, who are at the initiation stage of their relationship, their financial dependence might reduce their likelihood of leaving a relationship if there is violence.

Since many young women in schools are not working, the opportunity to remain at school may mean that they cannot leave the relationship. Young women are likely to date partners who are able to buy food, clothes, and other things. Further, young women under such conditions accept the violence that is perpetrated by their partners and this includes having sex when they do not want to and performing other sexual activities. Thus, the young women’s reliance on partners, exposes them to a higher risk of IPV (Da Cruz et al., 2014). This understanding emphasizes the public health burden of IPV and its rationale among young women experiencing financial constraints in a developing country such as Mozambique. There is indication of an initiative aimed at improving the economic and financial status of women in Mozambique, through promotion of entrepreneurship and employment opportunities, as an element of the 2030 Sustainable Development Agenda. However, the reported “National Program for women’s economic empower” was only launched recently in 2019, and it prioritizes the illiterate women from rural areas and is not addressing the needs of school going young women [56]. Notwithstanding this, there is a decade in which progress can be made towards the SDGs in Mozambique, and to develop targeted strategies towards the 2030 goals. The findings of this study contribute evidence-based information to advocate for specific prevention programs to improve the economic circumstances of young women in schools.

Community factors.

Religiosity. Our multivariable analysis confirmed that the odds of experiencing IPV increased substantially among women who do not consider themselves committed to any religion, in contrast to their counterparts (AOR, 1.596, 95% CI: 1.009–2.525, p = 0.046). Being a person who follows religious principles suggests that these women were less likely to be involved in risky behavior, which can lead to IPV, or may suggest that they are more likely to use nonviolent methods when in conflict situations. In contrast to these results, a qualitative study among Togolese women reported a high tolerance of IPV among those who were committed to religion [57]. This could be because in some religion, women are supposed to respect and obey their partner at all times, and IPV may be acceptable according to such religious beliefs. For example, the findings exploring the links between religious affiliation and IPV among women in Ghana, reported trends in sexual and emotional IPV among women adhering to a religion, when compared to those not involved with any religion affiliation. However, they were at the same time less likely to report physical violence [53]. However, in the Mozambican context, there are many different churches and people can freely choose to attend whichever they prefer. Therefore, people who commit themselves to religious principles are less likely to be involved in risky behavior, including sexual and physical violence. However, they may at the same time, be more likely to be tolerant of violent behavior if their religion enforces gender inequalities and ideas about male dominance and female submissiveness [58]. Hence, religiosity may even hinder the women recognizing their risk and reporting the abuse if the religion’s views accept violence and this may perpetuate IPV [59]. The differences reported in our results may be due the methods we used to assess religiosity. We assessed religious commitment as a yes/no responses, which could have been too crude to provide sufficient information regarding religiosity as a risk or protective factor for IPV. The role of religiosity and its relationship to IPV in this setting appears to be complex and calls for further research on how religious values might shape social attitudes and perspectives on IPV. The various religions could also have a role in preventing IPV, through their social and cultural perspectives.

Through collaboration with community members, leaders and the educational sectors, IPV awareness and preventions strategies could be discussed and implemented taking into consideration both the needs of individuals and the community’s religious values and the detrimental consequences of IPV on women and children.

Societal factors.

Young women’s beliefs about male superiority. In many traditional societies beliefs about male superiority remain strong, and the empowerment of women has proved to be a slow process (Shamu et al. 2015). Amongst the young women in our study despite their attending secondary school, in the bivariate analysis, a statistically significant association was found between experiencing IPV (OR 2.298, 95% CI: 1.014–5.210 p = 0.046) and their believing that “males are superior to females”. The possibility of changing such beliefs is feasible since our results differ from studies done among adolescent learners in South Africa. Young women there reported empowered attitudes, disagreeing with male superiority, and these attitudes were associated with a reduced risk of being emotionally abused [46, 47]. This reduction in abuse can be interpreted either as an indication that such women may obey their partner and not argue, or it may be hypothesized that young women who disagree with ideologies of male superiority are more likely to manage conflicts by resolving them within the relationship. In these studies, although the attitudes of women who disagree with ideas of male superiority reduced their risk of emotional violence, they remained at risk of physical violence, emphasizing the importance of including both sexes in school-based and community-based initiatives to reduce gender violence. Of concern is that both the female and the male learners in a South African studies, advanced ideologies of entitlement and male superiority [46, 47]. These perceptions are very traditional and are found in many cultures. Such beliefs need to be challenged in order to achieve more egalitarian societies which respect and develop the contribution of women, who often constitute half the population. Such an approach is essential to reduce IPV and achieve the SDG 5.

Consistency regarding the association between the sociocultural factors and the risk of IPV has also been reported from SSA and elsewhere [41, 54] and in studies conducted in different countries in low-middle income countries, where empowered women were less likely to report IPV [20].

Many statements regarding gender issues were not found to be associated with IPV in this study. Further, the multivariable logistic regression analyses did not confirm the trends of the bivariate analysis. The likely reason for these results could be the setting where the study was undertaken, since schools are often where young women receive education and health promotion programs and this population is possibly more aware of the required social norms and gender issues.

The current prevention programs on partner violence in Mozambique indicate collaboration with the educational sectors to promote gender equality [60]. However, the findings from our study revealed a surprising and concerning high prevalence of IPV and this appears to indicate a gap among young women in schools, regarding their skills to challenge male dominance norms and to effectively prevent IPV. Further, obtaining information regarding health promotion and how to prevent risky behaviors and violence, proved to be difficult among the school going women in our study, as they explained after answering the questionnaire. This suggests that the existing sociocultural vulnerabilities among young women may result from the social context or environment where they are integrated, such as the community where they live, which may endorse strong ideologies of male dominance, rather than the educational setting. In Mozambique, concurrent gender norms which privilege male dominance over women still exist. For example, in some cultures, it is perceived as normal and acceptable for men to have more than one partner [36, 61]. These findings were consistent with those of a previous study evaluating the multisectoral response of gender-based violence (GBV) in Mozambique, which revealed that young women justified that a man has a right to have sex and that women should satisfy men at any time [23]. Such views are also consistent with studies done in similar settings [13, 20, 48, 53, 54]. Similarly, social norms of male dominance and the risk of IPV against women were also described in recent studies from Bangladesh [10], Vietnam [9] and Nepal [8].

The finding from our study raise concern about the young women’s low gender empowerment attitudes and indicate an urgent need for the inclusion of education on gender equality early in the school programs and the importance of involving both sexes for effective results. Further a systematic review of studies conducted to evaluate preventive interventions on violence against adolescents and young girls in Lower Income Countries, including the SSA countries, pointed out that the persistent sociocultural factors may limit the success of prevention programs in these settings [4, 62].

There is thus a need for consideration of the contextual sociocultural environment and to include comprehensive programs that empower younger women with skills to challenge such negative gender norms when implementing IPV programs in schools. Comprehensive programs require collaboration with the community and the educational sector in working to change gender norms regarding male dominance. This can further improve social interactions and relationships and thus prevent violence and IPV.