Examining gender based violence and abuse among Liberian school students in four counties: An exploratory study

Child Abuse & Neglect 44 (2015) 76–86

Contents lists available at ScienceDirect

Child Abuse & Neglect

Examining gender based violence and abuse among Liberian school students in four counties: An exploratory study夽 Judy L. Postmus a,∗ , Gretchen L. Hoge a , Rebecca Davis b , Laura Johnson a , Elizabeth Koechlein c , Samantha Winter a a Center on Violence Against Women & Children, Rutgers University, School of Social Work, 390 George Street, Suite 408, New Brunswick, NJ 08901, USA b Center for International Social Work, Rutgers University, School of Social Work, 536 George Street, New Brunswick, NJ 08901, USA c United Nations Research Institute for Social Development, Palais des Nations, 1211 Geneva 10, Switzerland

a r t i c l e

i n f o

Article history: Received 11 July 2014 Received in revised form 28 October 2014 Accepted 4 November 2014 Available online 16 December 2014 Keywords: Gender-based violence Sexual violence School sample Liberia Transactional sex Sexual coercion

a b s t r a c t The purpose of this article is to uncover the extent of sexual gender based violence (GBV) experienced by a convenience sample of students from select counties in Liberia and to understand the disclosure experiences of those victims willing to come forward. Girls (n = 758) and boys (n = 1,100) were asked about their sexual GBV experiences including their disclosure experiences, if applicable. Results indicated that sexual violation (i.e., peeping or inappropriate touching) was found among both girls and boys. Sexual coercion (i.e., forced sex) was more prevalent than transactional sex (i.e., trading sex for grades or money). Both sexual coercion and transactional sex were reported by more girls than boys, yet the rates for the most severe form of sexual violence (i.e., sexual coercion) were high for both girls (30%) and boys (22%). When students were asked if they told anyone, 38% reported that they did disclose their experiences. This study contributes to a small but growing body of research to document the prevalence and types of sexual violence against children in Liberia. Consistent with other studies, the evidence shows that sexual violence against boys and girls is occurring at alarming rates. © 2014 Elsevier Ltd. All rights reserved.

Gender based violence, as it is currently experienced by women and children in Liberia, takes place against a historical backdrop of large-scale struggle for power and violent conflicts. Liberia’s historically hierarchical social structure is reflected in the status of “Americo Liberians,” those who settled either as ex-slaves or free persons from the U.S. in the 1820s. This group is also disproportionately represented in the past heads of state and held a significant proportion of wealth and power until the 1980s. The occurrence of military coups in 1980 by Samuel Doe, from the Krahn tribe, and Charles Taylor with the National Patriotic Front of Liberia (NPFL) in 1989 were followed by corrupt and violent regimes that eventually led to civil war that inflicted immense atrocities against civilians. Despite a brief peace after Taylor’s presidential election in 1997, violent conflict broke out again from 1999 to 2003. The civil war in Liberia was brutally violent and often included the use of child soldiers, torture, and rape as a weapon. Half of all Liberians were forced to flee their homes during the civil conflict (Swiss et al., 1998) and two million people were displaced. Kenneth Cain (1999) notes that the first conflict resulted in between 200,000 and 270,000 casualties, 750,000

夽 This project was supported by Ibis as part of the Liberian Research Consortium on Gender Based Violence. Points of view in this document are those of the authors and do not necessarily represent the official position or policies of Ibis or members of the Research Consortium. ∗ Corresponding author. http://dx.doi.org/10.1016/j.chiabu.2014.11.012 0145-2134/© 2014 Elsevier Ltd. All rights reserved.

J.L. Postmus et al. / Child Abuse & Neglect 44 (2015) 76–86

77

international refugees, 1.2 million internally displaced persons, and thousands more people who were tortured, maimed, and otherwise victimized. Rape was also used as a weapon during the conflict (Dziewanski, 2012), and while estimates for all vary, rates of violence against both combatants and civilians was high. A review of studies using population-based surveys estimates that 10–20% of Liberian women were raped during the Liberian civil war (Cohen & Green, 2012). President Ellen Johnson-Sirleaf was inaugurated as the first female president in post-civil war Liberia and on the African continent in 2006. She has attempted to shape her presidency around “gender sensitive reforms and the absence of armed conflict” (Popovic, 2009, p. 9), and the Liberian government has set gender equality and the elimination of violence high on its agenda. Despite these positive changes, rebuilding Liberia has proven challenging due to its violent recent history. Not only destructive to the social fabric of communities, families, and the psyche of its inhabitants, the conflict temporarily plunged Liberia into failed state status as a result of the economic and political consequences of conflict that resulted in higher levels of poverty and reduced access to services. Women in Liberia, already challenged by patriarchal social structures and institutions not reflective of their interests in access to employment, education, marriage and landownership, found themselves even more vulnerable during the conflict. Political and economic instability post-conflict has exacerbated this struggle. As Liberia strives to rebuild, efforts are underway to address gender based violence (GBV) throughout the country, especially violence against women and children. Understanding the scope and nuances behind such victimization, especially among children, is a first step to developing policies and programs to address and prevent such violence; however, much of the literature related to sexual violence and GBV in Liberia is focused on violence during conflict. The purpose of this article is to uncover the extent of sexual violence experienced by a convenience sample of students from select counties in Liberia and to understand the disclosure behaviors of those victims willing to come forward. Gender Based Violence in Liberia The extent of sexual violence during and following the Liberian civil war is debated and difficult to determine. Cohen and Green (2012) note that estimates by various human rights organizations and media during and immediately after the conflict indicated that approximately two-thirds of women were raped during the conflict. However, they point out that the studies used to obtain these numbers were often not designed to evaluate the overall prevalence of rape during the war, but rather for the purpose of understanding the experiences of those who suffered and reported rape. More reliable data suggest that the rate was between 10 and 20% of women experienced sexual violence during the civil conflicts (Cohen & Green, 2012). One nationally based study was conducted by the United Nations Mission in Liberia (UNMIL) in which researchers randomly selected 1,000 households in all fifteen counties in Liberia to determine the prevalence of and attitudes toward rape (UNMIL, 2008). Out of the total 2,952 participants who were interviewed, 37.3% were male while 67.2% were female of 10 years of age and above. Approximately 22% of respondents knew someone who had been raped, while 2% indicated that they had been raped. Rape survivors were generally between the ages of 10 and 19 while 41% of perpetrators were between 20 and 39 years old (UNMIL, 2008). Unfortunately, the researchers relied on a narrow definition of rape as dictated by Liberian law and a focus on penetration only. Additionally, the researchers reported that some parents responded on behalf of their children. Finally, the researchers identified that respondents were not willing to talk about rape or even sex but instead, only talked about incidents experienced by others (UNMIL, 2008). A more recent population-based study on gender based violence among 600 adult females in two Liberian counties (Montserrado and Nimba counties) found 56–59% of females experienced at least one act of gender based violence over the past 18 months (Stark, Warner, Lehmann, Boothby, & Ager, 2013). However, another population-based study that surveyed 4,501 adults across the country found that almost 38% of women and over 15% of men had been severely beaten by a spouse or partner during their lifetime (Vinck & Pham, 2013). Unfortunately, both studies only provide information on adult populations with a more general view of violence that is primarily focused on physical violence between partners. Additionally, though both studies used rigorous methods, the results of experiencing intimate partner violence varied quite a bit from 56–59% (more recent violence) to as low as 38% (lifetime violence). In another study on intimate partner violence among adult women, Horn, Puffer, Roesch, and Lehmann (2014) conducted 14 focus groups with 110 women from Sierra Leone and Liberia regarding their experiences with intimate partner violence post-war. The study found intimate partner violence to be an area of concern for women in both countries and factors such as cultural beliefs surrounding gender norms and women’s financial dependence on men contribute to the prevalence of intimate partner violence. Some women agreed that women’s increased involvement in their communities and the presence of NGOs following the war contributed to a decrease in intimate partner violence in some situations (Horn et al., 2014). After the war, there were more economic opportunities for women, which provided greater financial independence. Additionally, women became more knowledgeable about their legal rights and services available for survivors of gender based violence (Horn et al., 2014). In Liberia, even though rape is a crime, fear, isolation, economic hardship, and mistrust of authority mean that many instances of gender based violence go unreported and the perpetrators are not identified, caught, or convicted (Ackerman, 2009). Only 6% of all reported cases of gender-based violence in 2010, which includes sexual violence or physical violence, made it to court with only 2% of the cases resulting in a conviction (Dziewanski, 2012). This is linked to both limited capacity of the law enforcement and justice systems as well as underreporting (Dziewanski, 2012). Many women do not report

78

J.L. Postmus et al. / Child Abuse & Neglect 44 (2015) 76–86

rape or sexual violence due to fear of stigmatization by the community as well as the expense required to travel to the police stations, mainly concentrated in cities in the most populated counties (de Carvalho & Nagelhus Schia, 2011). The United Nations Mission in Liberia (UNMIL, 2008) has also reported that 65% of victims told someone about the rape and 16% reported the rape to the police. The survey of attitudes revealed that 81% of respondents exhibited rape supporting attitudes, such as feeling the victims contributed to their victimization while other respondents felt poverty, war, and unemployment were factors that caused rape. More than half of the respondents, 56%, believed that women lie about rape (UNMIL, 2008). Despite these challenges, Medie (2013) suggests that the rate of rape cases referred by the police to court has increased from 1967 to 2009. Medie (2013) attributes this in part to the participatory role that Liberian women have had in activism following the Second Liberian War. In collaboration with non-governmental organizations, international organizations and state agencies, the Liberian women’s movement has advocated for amendments to pre-existing rape laws, which has worked to expand the legal definition of rape and impose harsher penalties for those found guilty. In addition, advocates have focused on educating the public and law enforcement officials on how community members should respond to rape victims (Medie, 2013). GBV in Liberian Schools GBV has also been found to occur in Liberian schools as a result of the economic realities faced by students and their families. This has meant that even when girls have the freedom to attend school instead of earning a livelihood or tending to other family responsibilities, school fees and materials can still be prohibitively high. Some have suggested that expenses for things like books, pencils, uniforms and shoes leave girls to use their bodies as currency (Ferris, 2007). In a recent study, researchers conducted focus groups with girls and boys in sixth and seventh grade in Monrovia, and found that both reported transactional sex for grades or school fees exchanged between girls and older men in schools (Atwood et al., 2011). It should be noted that although the students were all in sixth and seventh grade, their ages varied widely due to many over-aged students in all grades. Male participants commonly viewed money as a means to acquire girlfriends while girls reported pressure to engage in transactional sex, or “Man-Man Business,” for status and money. Often these men were either teachers or men in other positions of authority (Atwood et al., 2011). Transactional sex between young girls and older men was prevalent enough that many of the children interviewed reported that it was at times encouraged by parents because it meant access to education that might otherwise be impossible (Atwood et al., 2011). Additionally, many of the girls who engaged in transactional sex reported pressure to do so from their peers, as it enhanced their social prestige. It was reported that teachers and parents did not always see the unequal power dynamics in a sexual relationship between young girls and older men as an issue and often ignored and normalized the violence against young girls (Atwood et al., 2011). Furthermore, young girls were frequently blamed for violence perpetrated against them through negative stereotyping. A qualitative study conducted by IBIS in rural southeast Liberia found that community members believed that girls (aged between 15 and 24) were pursuing materialistic gain through marriage, as weak or lazy, or focused only on domestic roles (Gronborg Helms, 2009). Once a girl reached physical maturation, regardless of her age, she was often sent to ‘look for support’ for her family through transactional sex, or a boyfriend who could support her financially. Having a ‘boyfriend’ at a young age was seen as fashionable by some and as a way of making ends meet for others. Also prevalent in the IBIS study was the reported feeling that girls were getting ‘too big, too soon’ and in a ‘rush to crush,’ placing the blame for having ‘boyfriends’ on young girls (Gronborg Helms, 2009). These views relieved potentially complicit parents of responsibility, and absolved the ‘boyfriends’ of their culpability in manipulating girls in exchange for money or material goods. To address the problem of GBV against children in schools in Liberia, a number of Liberian ministries and local international non-governmental organizations (NGOs) formed a Research Consortium. Consortium members include representatives from the following organizations and agencies: Ministry of Education; Ministry of Gender and Development; Association of Liberian Universities; IBIS; The Norwegian Refugee Council; Save the Children; and Concern Worldwide. After discussions spanning a year, the Consortium invited applications for a preliminary and short-term project to collect empirical data on the prevalence of GBV and its impact on girls. The invitation required a research partnership between a university in Liberia and an external university. This article outlines the project, funded by this Research Consortium, which assessed gender based violence (GBV) against youth in Liberian schools in four counties including Montserrado, Bong, Grande Gedeh and Grand Bassa. The research questions driving the study included: (1) What types of violence did students experience? (2) Did students disclose these experiences? Is so, to whom did they disclose? If not, why not? (3) Did certain types of violence relate to disclosure or not? (4) Were there any differences in experiences and disclosure based on gender? Methods Preliminary research and data collection were carried out in three phases in 2012. Phase I included reviewing and analyzing existing policies, laws, research reports and other documents that address GBV specific to Liberia. Phase II, which ran concurrently with Phase I, included the identification and interviewing of key stakeholders including Consortium members, representatives of international and local NGOs and representatives of key ministries and departments (e.g., the Ministry of Education of Liberia and the Ministry of Gender and Development). Phase III involved field research conducted by the Liberian

J.L. Postmus et al. / Child Abuse & Neglect 44 (2015) 76–86

79

Table 1 Sample location. County Bong Montserrado Grand Gedeh Grand Bassa

Overall (n = 1,858) 468 596 390 404

Male (n = 1,100) 256 319 270 255

Female (n = 758) 212 277 120 149

research team at selected schools in four counties in Liberia. Data collection included a mix of qualitative and quantitative methods. This article focuses on quantitative data gathered through an anonymous student survey during Phase III. Data Collection – Anonymous Student Survey Four counties in Liberia were selected and included Montserrado, Bong, Grande Gedeh and Grand Bassa counties. These counties were chosen based on input from our research and funding partners based on their connections to the school districts in these counties. The Liberian research team used non-random convenience samples within each county to select at least four schools in the four counties of interest with input our funding partners. Four schools were selected in Bong, Montserrado and Grand Bassa counties, while six schools were selected in Grand Gedeh County. The number of counties and schools selected were determined by the level of funding and the timeframe for the study. All schools selected agreed to participate in the study. The anonymous student survey was administered to all female and male students in the selected schools and contained sixty-eight items comprising different categories related to GBV. Students were organized by gender and grade level for participation in the survey with no students under the age of ten allowed to participate. Survey questions were read aloud in English by the researchers while the students marked their answers on answer sheets, which were turned in to the researchers at the conclusion of the survey and kept confidential. Sample In total, 2,166 students participated in the survey, including 1,236 boys and 930 girls. Listwise deletion was used to remove any case with incomplete data for questions related to GBV experiences from the overall analysis. The remaining overall base analytic sample included 1,858 participants, with 1,100 boys and 758 girls. Table 1 shows a breakdown of participants by county and gender. Protection of Human Subjects Prior to initiating each phase of the study, all materials were submitted to the Institutional Review Board (IRB) at the U.S. University. Unfortunately, there was no such board at our partnering Liberian university; however, the project and all personnel involved with the study were approved by the Vice President of Research (research authority) at the Liberian university. IRB approval ensured that research participants were informed of their rights, assured of confidentiality and anonymity, and that participants would not experience adverse effects as a result of participating in the study. All study participants were informed that their participation was voluntary and that they could withdraw or refuse to answer any particular question at any time. Liberian research assistants obtained verbal assent from all students prior to participating in the anonymous student survey. Additionally, consent for the anonymous student survey was provided by the County Education Officers (CEOs), the District Education Officers (DEOs) and school principals. Measures Survey questions were developed based on analysis of key stakeholder interviews from Phase II of the study and existing scales related to GBV and attitudes toward women. All questions were reviewed and revised by the Research Consortium and members of the Liberian university research team to ensure applicability to the local context and culture. Gender Based Violence Experiences. Initial analysis was run on 13 survey questions that focused on GBV experiences. Due to the presence of missing data and unclear contextual nature in which these actions could occur, two questions related to being shown (20.9% missing) or made to touch (19.8% missing) someone’s penis were removed from further analysis. The question, “Has someone ever touched you on your butt when you didn’t want it?” was also removed due to the unclear contextual nature in which this action could occur. Another question, “Has someone ever said something to scare you to do something you didn’t want to do?” was also removed due to its unclear link to sexual or gender based violence. The final measure of GBV experiences included nine questions. Each question asked participants if they had ever experienced a particular type of GBV. Response options for each item were 0 = no, 1 = yes.

80

J.L. Postmus et al. / Child Abuse & Neglect 44 (2015) 76–86

The nine questions included in the final measure of GBV experiences were then divided into three conceptual GBV categories for the purposes of analysis: Sexual Violation, Sexual Coercion, and Transactional Sex. Sexual Violation included three questions (“Watched you go to the bathroom or get dressed? [Peeped you?],” “Touched you on the breast when you didn’t want it? [Girls only]” and “Touched you on the penis? [Boys only]”). The item “Watched you go to the bathroom or get dressed? (Peeped you?)”, was added to the survey by our Liberian colleagues. This particular question is problematic in that it does not distinguish intention or motivation behind the viewing; nor does it indicate whether these behaviors were a normal part of family life or unwanted. However, we wanted to capture the full range of possible sexual violence experiences from the more “covert” and “harmless” to the more overt and “obvious” forms of violence. The data collectors were trained to instruct students of the purpose of this study, its intent to identify gender based violence, and the inclusion of “peeped you” as a form of unwanted sexual behavior. Sexual Coercion included three questions (“Threatened you with a knife, gun or other weapon to force you to have sex?” “Threatened to hurt you because you refused to have sex?” and “Forced you to have sex when you didn’t want to?”). Transactional Sex included three questions (“Asked you for sex to get a better grade?” “Asked you for sex to give you money or help?” “Asked you for sex to give you uniforms or food or something similar?”). Participants who responded “1 = yes” to at least one individual item in a conceptual category of GBV received a score of “1 = yes” for that particular category. If the participant did not respond “1 = yes” to at least one individual item in a conceptual category, the participant received a “0 = no” score for that particular category. Disclosure Behaviors Related to Gender Based Violence. Disclosure behaviors related to GBV were explored through various measures. The first measure was a single question asking those participants who had experienced GBV whether or not they had told anyone. Response options for this question were 0 = no, 1 = yes. The second measure related to disclosure of GBV experiences asked participants who disclosed to someone about their GBV experiences to identify to whom they disclosed. This measure included nine questions offering distinct options for disclosure (e.g., friend, parent, family member). Participants were asked to respond 0 = no, 1 = yes to each question. The third measure related to disclosure of GBV experiences asked participants who had experienced GBV but who did not disclose this information to anyone to identify the reasons why they did not disclose. This measure included seven questions, each providing a distinct reason for non-disclosure (e.g., “I was afraid of what would happen to me”). Participants were asked to respond 0 = no, 1 = yes to each question. Demographics. For this study, gender was self-reported by students through the anonymous survey. Response options were 1 = Male, 2 = Female. We purposefully did not ask the ages of the respondents in order to keep their answers completely anonymous. Since there is a wide range of ages in each class, it is conceivable that the data collectors would be able to match answers to a particular respondent who might be the only person in that class that was that age. Analysis Strategy Descriptive statistical analyses were run including frequencies on all individual survey items and sum and mean scores for measures of GBV experiences and disclosure behaviors. Disclosure behaviors were also examined by GBV experience categories via descriptive statistics. Chi square analysis examining differences based on gender for individual items and categories of GBV experiences are based on this overall sample of 1,858 (1,100 boys, 758 girls). In an effort to report findings on as many participants as possible, pairwise deletion was used for chi square analyses examining differences based on gender for examining to whom participants disclosed and why participants did not disclose. The use of pairwise deletion in these analyses resulted in varied sample sizes for each item that are indicated in each results table. The base analytic sample for analysis of to whom participants disclosed included 708 participants (421 boys, 287 girls) who had experienced GBV and disclosed this information to someone. The base analytic sample for analysis of reasons why participants did not disclose included 1,150 participants (679 boys, 471 girls) who had experienced GBV but had not disclosed this information to anyone. Listwise deletion was used for examining categories of GBV experienced and to whom participants disclosed. Since GBV categories were not mutually exclusive, sample sizes for each type of GBV experienced are as follows: Sexual Violation (n = 436), Transactional Sex (n = 221), Sexual Coercion (n = 287). Listwise deletion was also used for examining categories of GBV experienced and why participants did not disclose. Again, as categories of GBV experiences were not mutually exclusive, sample sizes for each type of GBV in this analysis are as follows: Sexual Violation (n = 683), Transactional Sex (n = 223), Sexual Coercion (n = 303). Logistic regression analysis was also conducted to determine if having experienced a particular type of GBV was related to disclosure behavior and whether gender moderated this relationship. Listwise deletion, resulting in an analytic sample of 1,690, was used for the logistic regression analysis examining the moderating effect of gender on the relationship between categorized type of GBV experienced and dichotomous disclosure of GBV experiences.

J.L. Postmus et al. / Child Abuse & Neglect 44 (2015) 76–86

81

Table 2 GBV experiences (% yes). Has someone ever. . . Sexual Violation Watched you go to the bathroom or get dressed? (Peeped you?)* Touched you on your breast when you didn’t want it? (Girls only) Touched you on your penis when you didn’t want it? (Boys only) Transactional Sex* Asked you for sex to get a better grade?* Asked you for sex to give you money or help?* Asked you for sex to give you uniforms or food or something similar? Sexual Coercion* Threatened you with a knife, gun or other weapon to force you to have sex? Threatened to hurt you because you refused to have sex?* Forced you to have sex when you didn’t want to?* At least one GBV experience Overall mean score for GBV experiences*

Overall (n = 1,858) 86.6 54.1 N/A N/A 32.1 15.1 22.9 15.2 42.9 18.8 20.5 25.3 91.7 2.51 (1.81)

Male (n = 1,100) 85.2 59.7 N/A 76.4 26.5 13.2 17.3 14.2 40.1 20.2 17.4 22.3 91.0 2.41 (1.73)

Female (n = 758) 88.7 45.9 83.4 N/A 40.2 17.8 31 16.8 47 16.9 25.1 29.8 92.6 2.67 (1.92)

Note: Summary scores for Sexual Violation, Transactional Sex, and Sexual Coercion represent the percent of respondents who experienced at least one kind of GBV listed under that category. * Significant between boys and girls at p < .01 level.

Results Gender Based Violence Experiences All items related to GBV experiences were answered by the total sample of 1,100 boys and 758 girls. Almost 92% of study participants experienced at least one form of GBV. This included 91% of boys and 92.6% of girls. The average score out of eight possible GBV experiences for girls was 2.67 (SD = 1.92) whereas boys had an average of 2.41 (SD = 1.73) experiences. There were significant differences between responses from boys and girls on the following items for questions regarding experience with GBV. Almost 46% of girls and 60% of boys said that someone had ever watched her or him go to the bathroom or get dressed (i.e., peeped them), although it was not specified whether this was in a sexualized way or not. Almost 18% of girls and over 13% of boys said they had been asked by someone for sex to get a better grade. Thirty-one percent of girls and 17% of boys said they had been asked by someone for sex in return for money. Over 25% of girls and 17% of boys said that someone had ever threatened to hurt her or him because she or he refused to have sex. Almost 30% of girls and over 22% of boys said they had been forced by someone to have sex when she or he didn’t want to (Table 2). Acts of GBV were also separated into three categories according to characteristics and the nature of the act. These categories included Sexual Coercion, Sexual Violation, and Transactional Sex. Almost 87% of the overall sample reported having experienced at least one act of sexual violation. There were no statistically significant differences between the percent of boys and girls who had experienced sexual violation. However, statistically significant differences did exist between the responses of boys and girls for sexual coercion and transactional sex. Forty-seven percent of girls and 40% of boys said they had experienced at least one act of sexual coercion. Over 40% of girls as compared to 27% of boys experienced at least one act of transactional sex. Disclosure Behaviors Related to Gender Based Violence To Whom Participants Disclosed by Gender. Approximately 38% of girls and boys told someone about their GBV experiences. Of those participants that told someone, the majority told a friend about their GBV experiences (61%). Over 53% of the overall sample told a parent about their GBV experience(s). There were several significant differences between girls and boys regarding who they told, including telling a family member (57% of girls; 41% of boys); doctor, nurse or other medical professional (44% of girls; 32% of boys); teacher (43% of girls; 31% of boys); and police officer or other legal professional (29% of girls; 28% of boys). In sum, girls told an average of 3.7 (SD 2.82) people about their experience(s) with GBV, whereas boys told an average of 3.06 (SD 2.54) people about their experience(s) with GBV (Table 3). To Whom Participants Disclosed by Type of GBV. Almost 66% of the participants who had experienced sexual violation told a friend about their experience(s) with GBV. Over 55% of participants who had experienced an act(s) of sexual violation told a parent about their experiences. Forty-seven percent of participants told a family member about their experience(s). Over 37% of the participants told a doctor, nurse, or other medical professional about their experience(s) with sexual violation. Over 36% of participants told a teacher about what had happened; almost 32% told an unspecified member of the school staff. Almost 34% of participants told a town chief, community head or zone leader about the event(s); over 31% of the participants told a minister or other religious leader. Finally, over 34% of participants told a police officer or legal professional about the sexual violation incident(s) (Table 4). Over 72% of participants who had experienced an act(s) of transactional sex told a friend about their experience(s), whereas 61% told a parent, and 56% told a family member about their experience(s). Almost 48% of participants told a

82

J.L. Postmus et al. / Child Abuse & Neglect 44 (2015) 76–86

Table 3 To whom students disclosed their GBV (% yes). Overall (n = 708) Friend Parent Family member* Doctor, nurse or other medical person* Teacher* Town chiefs, community heads, zones leaders Police officer or other legal person* School staff Minister or religious leader Overall mean score*

Male (n = 421)

61 53.4 46.5 36.3 35.4 32.7 32.1 30.3 29.9 3.29 (2.66)

Female (n = 287)

57.7 53.3 40.8 32.3 31.3 29.4 28.2 27 27.2 3.06 (2.54)

67 53.5 57.4 43.8 42.9 38.6 39.2 36.3 34.8 3.7 (2.81)

The n value represents those participants who experienced at least one act of GBV and who said they told someone. * p < .01. Table 4 Type of GBV experienced and to whom participant disclosed (% yes). Sexual Violation (n = 436) Friend Parent Family member Doctor/nurse/medical person Teacher School staff Town chief/community head/zone leader Minister/religious leader Police officer/legal person

Transactional Sex (n = 221)

65.8 55.3 47.0 37.2 36.2 31.9 33.5 31.4 34.2

Sexual Coercion (n = 287)

72.4 61.1 56.1 47.5 45.2 40.7 42.1 38.9 40.3

70.4 59.2 54.4 44.6 40.8 36.2 40.4 35.9 40.1

The n value represents those participants who experienced that particular type of GBV and had no missing data on any of these variables.

doctor, nurse, or other medical professional about the event(s) of transactional sex. Over 45% of participants told a teacher about the transactional act(s) of GBV that had been committed against them; almost 41% told an unspecified member of the school staff about what had happened. Over 42% of participants told a town chief, community head, or zone leader; over 40% told a police officer or other legal professional about what had happened. Finally, almost 39% told a minister or other religious leader about the transactional act(s) of GBV. For those who experienced sexual coercion and told someone about their experience(s), over 70% told a friend, over 59% told a parent, and over 54% disclosed what had happened to a family member. Almost 45% of participants told a doctor, nurse, or other medical professional about their experience(s) with sexual coercion. Almost 41% of participants told a teacher about their experience(s) with sexual coercion; over 36% told an unspecified member of the school staff about what had happened. Over 40% of participants told a town chief, community head or zone leader about the event(s). Over 40% of participants told a police officer or other legal person. Finally, almost 36% told a minister or religious leader about what had happened. Reasons for Non-Disclosure by Gender. Those students who did not disclose their GBV experiences were also asked why they did not tell anyone. Over 43% of boys and girls who experienced GBV but did not report it did not think anyone would believe them. Over 41% were afraid of what would happen to them if they did report or were afraid of getting into trouble. Almost 38% were afraid of what would happen to their family. Almost 35% were afraid of getting kicked out of school. There were statistically significant differences in the responses of boys and girls on the following items including being afraid of the person who committed the GBV act (43% boys; 33% girls), and being ashamed of what had happened (46% boys; 37% girls) (Table 5).

Table 5 Reasons for not disclosing GBV experiences (% yes). Overall (n = 1,150) I was afraid of what would happen to me. I was afraid of what would happen to my family. I was afraid of getting into trouble. I was afraid I would get the person who did it to me in trouble.* I didn’t think anyone would believe me. I was afraid I would get kicked out of school. I was ashamed of what had happened and didn’t want anyone to know.*

41.2 37.8 41.4 38.7 43.3 34.5 42.3

Male (n = 679) 41.3 38.5 45.3 43.2 45.4 35.1 46.2

The n value represents those participants who experienced at least one act of GBV and who said they did not tell anyone. * Significant between boys and girls at p < .01 level.

Female (n = 471) 41.2 36.9 36.3 32.8 40.4 33.8 37.2

J.L. Postmus et al. / Child Abuse & Neglect 44 (2015) 76–86

83

Table 6 Type of GBV experienced and reasons for not disclosing (% yes). Sexual Violation (n = 683) I didn’t think anyone would believe me Afraid of what would happen to me Afraid of what would happen to my family Afraid of getting into trouble Afraid I would get the person who did it to me in trouble Afraid I would get kicked out of school Ashamed of what had happened and didn’t want anyone to know

Transactional Sex (n = 223)

Sexual Coercion (n = 303)

43.3 41.6 39.2 43.2 39.5

44.8 44.4 42.2 43.0 37.7

52.5 51.2 45.2 47.5 43.6

34.6 43.5

42.6 45.3

43.9 50.2

The n value represents those participants who experienced that particular type of GBV and had no missing data on any of these items. Table 7 Binary logistic regression of disclosure on gender on type of GBV (n = 1,690). Odds ratios

Confidence interval (95%) Lower bound

Gender Sexual Coercion Sexual Violation Transactional Sex

0.852 1.710*** 1.381 1.641***

0.690 1.377 0.988 1.307

Upper bound 1.051 2.124 1.932 2.059

R2 = 0.057. The n value represents those participants who experienced any type of GBV and had no missing data on the items used for analysis. Significance based on two-tailed T-test: *p < 0.05; **p < 0.01; ***p < 0.001.

Reasons for Non-Disclosure by Type of GBV. For those participants who experienced an act(s) of sexual violation, over 43% did not tell anyone about their GBV experiences because they did not think anyone would believe them, were ashamed of what had happened and didn’t want anyone to know, or were afraid of getting into trouble. Almost 42% did not tell anyone because they were “afraid of what would happen to me” (the victim). Over 39% did not tell anyone because they were afraid of what would happen to their family; 43% did not tell for fear that the perpetrator would get in trouble. Almost 35% of participants were afraid of getting kicked out of school (Table 6). For those participants who experienced an act(s) of transactional sex, over 45% of participants were ashamed of what had happened and didn’t want anyone to know. Almost 45% didn’t think anyone would believe them. Around 44% did not tell anyone about their GBV experiences because they were afraid of what would happen to them (the victim). About 43% were afraid of getting kicked out of school or of getting into trouble. Over 42% were afraid of what would happen to their family and almost 38% were afraid that they would get the perpetrator in trouble. For those participants who experienced an act(s) of sexual coercion, almost 53% of participants did not think anyone would believe them. Over 51% did not tell anyone because they were afraid of what would happen to them (the victim). Almost 48% of participants were afraid of getting into trouble, over 45% were afraid of what would happen to their family, almost 44% were afraid of getting kicked out of school or were afraid that they would get the perpetrator in trouble. Over 50% of participants were ashamed of what had happened and did not want anyone to know. Type of GBV Experiences and Relationship to Dichotomous Disclosure Behavior Results from the logistic regression suggested that gender did not have a significant main effect on whether or not participants disclosed to someone about their GBV experiences. However, results suggested participants who experienced sexual coercion have 71% greater odds of disclosing GBV experiences compared to those who did not experience sexual coercion but who had experienced sexual violation and/or transactional sex, controlling for gender and other types of GBV experiences. In addition, participants who experienced transactional sex have 64% greater odds of disclosing GBV experiences compared to those who did not experience transactional sex but who had experienced sexual coercion and/or sexual violation, controlling for gender and other types of GBV experiences. Analysis was also run to determine the moderating effect of gender on the relationship between type of GBV experienced and disclosure of GBV experiences. However, no significant moderating effect was found (Table 7). Discussion The results from this study provide an initial glimpse into the problem of sexual violence experienced by girls and boys in schools in four counties in Liberia. Results indicated that sexual violation (e.g., peeping or inappropriate touching) was very prevalent among both girls and boys. Sexual coercion (e.g., forced sex) was more prevalent than transactional sex (e.g., trading sex for grades or money). Both sexual coercion and transactional sex were reported by more girls than boys, yet the rates for the most severe form of sexual violence (i.e., sexual coercion) were high for both girls (30%) and boys (22%).

84

J.L. Postmus et al. / Child Abuse & Neglect 44 (2015) 76–86

These findings are similar to other studies of child sexual abuse, which have found high prevalence rates for male study participants in Africa (Madu & Peltzer, 2000, 2001; Stoltenborgh, van Ijzendoorn, Euser, & Bakermans-Kranenburg, 2011). One possible explanation offered was that the absence of an adult male in the household may have left male children vulnerable to abuse by adult females (Madu & Peltzer, 2000, 2001). A study by McCrann, Lalor, and Katabaro (2006) noted that 11% of males in their study reported they were bribed to engage in an unwanted sexual experience, which may indicate that males are experiencing transactional sex as well. As such, it is possible that many single-parent families and orphaned children emerged as a result of the Liberian civil war. Economic hardship and the absence of adult caretakers may have left Liberian children vulnerable to abuse and exploitation by older adults. When students were asked if they told anyone, 38% reported that they did disclose their experiences. Most tell their friends (61%) followed by telling their parents (53%) which is similar to other studies that outline such groups in which victims confide (Malloy, Brubacher, & Lamb, 2013; Smith et al., 2000). In this study, girls were more likely to tell a family member, medical staff, teacher, or police than were boys. Such differences could be related to different reactions to victimization based on gender; further research is needed to determine if such gender differences to disclosure exist. Similar disclosure patterns also held up when examining the different types of GBV experienced. The higher rate of disclosure to formal sources may be the result of considerable investments in Liberia in the last 5–6 years to improve the system for reporting to health clinics and police. The Women and Children’s Protection Section has been created with female police officers trained in gender-based violence and good practices in interviewing women and children (de Carvalho & Nagelhus Schia, 2011). The raised awareness and improvement in the process of reporting potentially increased the confidence in those who care for children including parents and teachers in how these situations will be handled with children and, thus, encourage children to report. These results indicated some positive trends in the sense that many students are telling others; the challenge lies with how those “others” respond to disclosure of GBV. Such a challenge can be found in other countries in which victims seek out more “informal” sources of help rather than “formal” sources; however, most efforts to train responders seem to focus on the formal sources (e.g., teachers, law enforcement, medical staff). The study also provided results on the reasons for students not telling anyone about their sexual violence experiences. Most answers hovered between 35 and 43% of the students with minimal differences between girls and boys. Boys reported being more afraid of getting the person (i.e., perpetrator) into trouble and were more ashamed of what happened. Consistently, those that experienced the more severe sexual coercive behaviors reported more reasons why they did not disclose those behaviors followed by those who experienced transactional sex. Finally, the results indicated that gender did not play a role in whether one would disclose victimization to someone else. However, experiencing a certain type of victimization (i.e., sexual coercion or transactional sex) did increase the odds that the victim would disclose to someone else. Fortunately, for children who are asked for sex for something (transactional sex) or who are forced to have sex (sexual coercion), they are more likely to tell someone about their experiences and potentially receive help. Further research is needed to determine what occurs as a result of disclosing such experiences. Additionally, it is possible that experiencing sexual violation (i.e., being peeped or being touched on one’s breast or penis) is not even considered a crime and hence has become normalized by these children. Further research is needed to fully understand what happens to children and the perpetrators when sexual GBV is brought to light.

Limitations The results from this study provide interesting details of the sexual violence experiences of girls and boys in schools in Liberia. However, the results should be taken with a note of caution. The counties, schools, and students were not randomly selected but were a convenience sample based on access to schools and on an existing presence within the counties of the respective NGOs. Therefore, students’ understanding and experiences in schools assisted by international organizations are not representative of students across the county or country. For these reasons, the results must be interpreted with caution and cannot be generalized to other students, schools or counties in Liberia. Additionally, the survey used in this study was created based upon the expertise of the researchers with input from the research consortium. The researchers thought that the existing measures, which have been tested, some in other African countries and some in high income countries in other parts of the world, may not be applicable with this study population. Prior to the study, several potential survey items were removed in consultation with the Consortium and the Liberian university research team since it was determined that they might not be fully understood by the study sample. We also did not have the opportunity to pilot the instrument prior to collecting the data for this project. More research is needed to develop and test culturally relevant and age-sensitive tools to be used in future studies. Finally, since we purposefully did not ask the ages of the respondents in order to keep their answers completely anonymous, further research is needed to untangle the differences experienced from younger students (age 10–12) from older students. However, the intent was to explore the phenomena of sexual violence against girls and boys attending Liberian schools. The results provide a first step in testing research tools and providing evidence of such violence as well as provide recommendations for addressing GBV in Liberian schools.

J.L. Postmus et al. / Child Abuse & Neglect 44 (2015) 76–86

85

Implications and Conclusion This study contributes to a small but growing body of research to document the prevalence and types of sexual violence experienced by children in Liberia. Consistent with other studies, the evidence shows that sexual violence against boys and girls is occurring at alarming rates. Even though Liberia has established sound policies on gender equality and the elimination of violence, efforts to address protection and justice for victims of violence are limited. One of the main issues is that there are restricted means for recourse for victims in terms of protection and justice, and there are few to no consequences for perpetrators. In the current Liberian context, reporting and not-reporting come at a price. Often, the victims report only to people they know, rather than directly to authorities, in an effort to protect themselves from further victimization due to stigma and the risk of being blamed. Protection of one’s own family, and in some cases, a perpetrator with status and power, may be reasons for not reporting. Access to protection and justice for victims and their families is clearly needed in order to address sexual violence against children. Providing justice to victims and retribution to perpetrators may go a long way in encouraging other victims to report incidents of violence while also sending perpetrators the message that their actions have consequences. Protection and justice requires not just human and financial resources, but also a better understanding of the reasons sexual violence is occurring, including the role of cultural beliefs and attitudes. But the causes of such blatant violation of the rights of children at family and community levels are complex. In addition, the evidence suggests that economic insecurity and gender insensitivity within institutional contexts contribute to the risk of sexual violence, especially as it relates to transactional sex. To better understand the influence of complex social norms and institutional factors within the Liberian context, further study is needed. Given that sexual violence is a shared problem and a high agenda item in many other Sub Saharan African countries, borrowing tested qualitative and quantitative methods can strengthen research efforts and contribute to the development of effective prevention strategies. Although in the minority, some victims are reporting incidences of GBV and finding that the perpetrators are being held accountable. A positive deviance approach to research that identifies those individuals and families that do report incidences of GBV could contribute to a better understanding of the specific variables that influenced those who did report to someone within the formal system (e.g., health professional, teacher or police). Positive deviance research refers to the study of those individuals in risk situations that deviate from the normative behavior and result in more positive outcomes for these individuals (Marsh, Schroeder, Dearden, Sternin, & Sternin, 2004). In addition, a positive deviance approach applied to organizational and system reform, such as poorly performing health care systems, identifies and studies those deviant practices that result in exceptionally, high performance in an area of interest (Bradley et al., 2009). Studying those cases in which GBV is reported and the systems in which justice is achieved could assist in defining those factors that contribute to improved accountability within the respective context. Another positive change is that sexual violence against children is on the public policy agenda and there is considerable discourse on the violation of human rights relative to GBV within Liberia. Research can help target specific attitudinal and behavioral changes needed and guide the development of public education and awareness initiatives. Community-level approaches that engage boys and men, as well as girls and women, are vital for examining and understanding gender roles and responsibilities. Liberian women activists have played a key role in strengthening current policies in Liberia surrounding gender based violence. Global NGOs should aim to partner with Liberian women and integrate a human rights framework in ways that are informed by Liberia’s history and gender relations (Abramowitz & Moran, 2012). Incorporating good practice models that facilitate gender-sensitive services inclusive of psychosocial, health, and educational services delivered within communities are needed. Researching and testing existing good practice models that engage local traditional, religious, and NGO leaders, Parent Teacher Associations (PTAs), and local child protection committees and authorities can contribute to prevention and protection. Contextualized gender-sensitive approaches can support the development of schools and communities as the safe havens they are meant to be for children.

Acknowledgements The authors would like to thank their collaborators at Cuttington University who played a principal role in helping us craft our methods and in collecting the vast amount of data. We are in their debt.

References Abramowitz, S., & Moran, M. H. (2012). International human rights, gender-based violence, and local discourses of abuse in postconflict Liberia: A problem of “culture”? African Studies Review, 55(2), 119–146. Ackerman, R. (2009). Rebuilding Liberia, one brick at a time. World Policy Journal, 26(2), 83–92. http://dx.doi.org/10.1162/wopj.2009.26.2.83 Atwood, K. A., Kennedy, S. B., Barbu, E. M., Nagbe, W., Seekey, W., Sirleaf, P., Perry, O., Martin, R. B., & Sosu, F. (2011). Transactional sex among youths in post-conflict Liberia. Journal of Health, Population, and Nutrition, 29(2), 113–122. Bradley, E. H., Curry, L. A., Ramanadhan, S., Rowe, L., Nembhard, I. M., & Krumholz, H. M. (2009). Research in action: Using positive deviance to improve quality of health care. Implementation Science, 4, 25. Cain, K. L. (1999). The rape of Dinah: Human rights, civil war in Liberia, and evil triumphant. Human Rights Quarterly, 21(2), 265–307. Cohen, D. K., & Green, A. H. (2012). Dueling incentives: Sexual violence in Liberia and the politics of human rights advocacy. Journal of Peace Research, 49(3), 445–458. http://dx.doi.org/10.1177/0022343312436769

86

J.L. Postmus et al. / Child Abuse & Neglect 44 (2015) 76–86

de Carvalho, B., & Nagelhus Schia, N. (2011). Sex and gender-based violence in Liberia and the case for a comprehensive approach to the rule of law. Journal of International Relations and Development, 14, 134–141. http://dx.doi.org/10.1057/jird.2010.26 Dziewanski, D. (2012). Peace Without security: Violence against women and girls in Liberia (Small Arms Survey Issue Brief Number 3). Retrieved from Small Arms Survey website: http://www.smallarmssurvey.org/fileadmin/docs/G-Issue-briefs/Liberia-AVA-IB3.pdf Ferris, E. G. (2007). Abuse of power: Sexual exploitation of refugee women and girls. Signs: Journal of Women in Culture and Society, 32(3), 584–591. http://dx.doi.org/10.1086/510338 Gronborg Helms, M. (2009). A small study on girl’s education in Southeast Liberia. Monrovia, Liberia: Ibis: Education for Development. Horn, R., Puffer, E. S., Roesch, E., & Lehmann, H. (2014). Women’s perceptions of the effects of war on intimate partner violence and gender roles in two post-conflict West African countries: Consequences and unexpected opportunities. Conflict and Health, 8(12), 1–13. http://dx.doi.org/10.1186/1752-1505-8-12 Madu, S. N., & Peltzer, K. (2000). Risk factors and child sexual abuse among secondary school students in the Northern Province (South Africa). Child Abuse & Neglect, 24(2), 259–268. http://dx.doi.org/10.1016/S0145-2134(99)00128-3 Madu, S. N., & Peltzer, K. (2001). Prevalence and patterns of child sexual abuse and victim–perpetrator relationship among secondary school students in the Northern Province (South Africa). Archives of Sexual Behavior, 30(3), 311–321. Malloy, L. C., Brubacher, S. P., & Lamb, M. E. (2013). “Because she’s one who listens”: Children discuss disclosure recipients in forensic interviews. Child Maltreatment, 18(4), 245–251. http://dx.doi.org/10.1177/1077559513497250 Marsh, D. R., Schroeder, D. G., Dearden, K. A., Sternin, J., & Sternin, M. (2004). The power of positive deviance. British Medical Journal, 329, 1177–1179. McCrann, D., Lalor, K., & Katabaro, J. K. (2006). Childhood sexual abuse among university students in Tanzania. Child Abuse and Neglect, 30(12), 1343–1351. http://dx.doi.org/10.1016/j.chiabu.2006.05.009 Medie, P. A. (2013). Fighting gender-based violence: The women’s movement and the enforcement of rape law in Liberia. African Affairs, 112(448), 377–397. http://dx.doi.org/10.1093/afraf/adt040 Popovic, N. (2009). Women, peace and security in Liberia: Supporting the implementation of Resolution 1325 in Liberia. Santo Domingo, Dominican Republic: United Nations International Research and Training Institute for the Advancement of Women (INSTRAW). Retrieved from http://www.peacewomen.org/assets/file/instraw liberia implementation.pdf Smith, D. W., Letourneau, E. J., Saunders, B. E., Kilpatrick, D. G., Resnick, H. S., & Best, C. L. (2000). Delay in disclosure of childhood rape: Results from a national survey. Child Abuse & Neglect, 24(2), 273–287. http://dx.doi.org/10.1016/S0145-2134(99)00130-1 Stark, L., Warner, A., Lehmann, H., Boothby, N., & Ager, A. (2013). Measuring the incidence and reporting of violence against women and girls in Liberia using the ‘neighborhood method’. Conflict and Health, 7(20) http://dx.doi.org/10.1186/1752-1505-7-20 Stoltenborgh, M., van Ijzendoorn, M. H., Euser, E. M., & Bakermans-Kranenburg, M. J. (2011). A global perspective on child sexual abuse: Meta-analysis of prevalence around the world. Child Maltreatment, 16(2), 79–101. http://dx.doi.org/10.1177/1077559511403920 Swiss, S., Jennings, P. J., Aryee, G. V., Brown, G. H., Jappah-Samukai, R. M., Kamara, M. S., Schaack, R. D., & Turay-Kanneh, R. S. (1998). Letter from Monrovia: Violence against women during the Liberian civil conflict. Journal of the American Medical Association, 279(8), 625–629. http://dx.doi.org/10.1001/jama.279.8.625 UNMIL (United Nations Mission in Liberia). (2008). Research on prevalence and attitudes to rape in Liberia September to October 2008. Retrieved from http://www.stoprapenow.org/uploads/advocacyresources/1282163297.pdf Vinck, P., & Pham, P. N. (2013). Association of exposure to intimate-partner physical violence and potentially traumatic war-related events with mental health in Liberia. Social Science & Medicine, 77, 41–49. http://dx.doi.org/10.1016/j.socscimed.2012.10.026