African Journal of Reproductive Health, Vol. 11, No. 3, December, 2007, pp. 62-82
Coerced First Sex among Adolescent Girls in Sub-Saharan Africa: Prevalence and Context
Ann M. Moore1, Kofi Awusabo-Asare2, Nyovani Madise3,4, Johannes John-Langba3, and Akwasi Kumi-Kyereme2
1Guttmacher Institute, New York, NY, United
Code Number: rh07035
Coercive experiences at sexual debut have been shown to be associated with other sexual risks throughout the life course. Using nationally representative surveys from 12-19 year old girls in Burkina Faso, Ghana, Malawi, and Uganda collected in 2004, we examine the prevalence of sexual coercion at sexual debut among unmarried girls and its correlates. In Malawi, 38 percent of girls said that they were "not willing at all" at their first sexual experience followed by Ghana at 30 percent, Uganda at 23 percent and Burkina Faso at 15 percent. In-depth interviews collected in 2003 with the same demographic shows that there are four primary types of sexual coercion: forced sex; pressure through money or gifts; flattery, pestering, and threatening to have sex with other girls; and passive acceptance. The article concludes with the research and policy implications of these findings.
Premier rapport sexuel coercitif chez les adolescents en Afrique subsaharienne: Prévalence et contexte On a montré que les expériences coercitives au moment du premier rapport sexuel ont été liées à d'autres risques sexuels pour toute la vie. A l'aide des enquêtes à représentation nationale menée auprès des filles âgées de 12-19 ans au Burkina-Faso, au Ghana, au Malawi et en Ouganda en 2004, nous avons étudié la prévalence de la coercition sexuelle au moment du premier acte sexuel chez les filles célibataires et leurs corrélats. Au Malawi, 38% des filles ont affirmé qu'elles "ne voulaient pas du tout" au moment de leur premier expérience sexuelle, suivi par le Ghana à 30%, l'Ouganda à 23% et le Burkina-Faso à 15%. Des interviews en prodondeur recueillies en 2003 avec les mêmes données démographiques, montrent qu'il y a quatre types de coercition sexuelle: le rapport sexuel contraint; la pression à travers l'argent ou des cadeaux; la flatterie, à force d'insister et menacer d'avoir des rapports sexuels avec d'autres filles; et l'acceptation passive. L'article conclut avec les implications de la recherche et de la politique de ces résultats.
Key Words: sexual coercion, sexual debut, girls
Worldwide, 40 percent of new cases of HIV infection occurred among young people between the ages of 15 and 24 years in 2006.1 Sub-Saharan Africa is the worst-affected region accounting for two-thirds of the HIV-positive population. Among those infected in sub-Saharan Africa, 59 percent are female.1 Unintended pregnancy is also a major reproductive health problem among young people in sub-Saharan Africa.2 In a number of countries, females aged 15-19 years account for between 10 and 15 percent of annual total fertility.3 Available evidence suggests that not all exposure to the risk of HIV infection and unintended pregnancy in sub-Saharan is voluntary.4-6
Sexual coercion is defined as "an individual woman's lack of choice to pursue other options [to avoid sexual interactions] without severe social and physical consequence."7 It ranges from deception to obtain sex to rape. What is experienced as coercion is gender- and culturally-specific. The relationship of the individuals and the circumstances under which the sexual intercourse occurs can impact individuals' perceptions of what constitutes sexual coercion. For example, youths in Nigeria stated that if a boy has spent money on a girl, then pressure from the male to engage in sex was acceptable.8
A growing body of research has reported significant associations between coerced sex and a range of negative reproductive and as well as psychological and emotional health outcomes. Reproductive health risks correlated with sexual coercion include sexually transmitted infections (which can cause cervical cancer and infertility) including HIV, unintended pregnancy which can possibly lead to unsafe abortion and as a consequence morbidity and even mortality, as well as the onset of risk-taking behaviors including other nonconsensual sexual experiences, multiple partnerships and unprotected sex.7;9-14 One study found evidence of sexual abuse victimization associated with later forcing someone else to have sex.15 A negative sexual experience can also result in a host of negative psychological outcomes including sexual dysphoria, anxiety, eating disorders, substance abuse, depression and even suicide or attempted suicide.16 A study in a township in Cape Town, South Africa found that women who had experienced coercion were significantly more likely to exchange sex for material needs, have multiple sex partners, engage in high rates of unprotected vaginal intercourse, and have more sexually transmitted infections (STIs).17 A recent study in Uganda found that adolescents who reported coerced first intercourse were significantly less likely than those who did not experience coercion at debut to be currently using modern contraceptives, to have used a condom at last intercourse and to have used condoms consistently in the last six months and were more likely to report their most recent pregnancy as unintended (among ever-pregnant women) and to report one or more genital tract infections.18 Fear of violence can also impact a woman's willingness to take protective action at the time of sex. A study among young women in South Africa found that partner violence and the fear of violence prevented girls from saying "no" to sex and compromised condom use.6
This paper draws on results from in-depth interviews and nationally representative surveys to examine coercive experiences at sexual debut of females aged 12-19 years in Burkina Faso, Ghana, Malawi and Uganda. The data are part of a study on adolescent sexual and reproductive health called Protecting the Next Generation: Understanding HIV Risk among Youth (PNG). The project seeks to contribute to the global fight against the HIV/AIDS epidemic among adolescents by raising awareness among young people's sexual and reproductive health needs with regard to HIV/AIDS, other STIs and unwanted pregnancy; communicating new knowledge to a broader audience including policymakers, healthcare providers and the media, in each country as well as regionally and internationally; and stimulating the development of improved policies and programs that serve young people. The larger study encompasses focus group discussions with 14-19 year olds; in-depth interviews with adolescents 12-19 years old, in-depth interviews with key adults (teachers, parents, community leaders and health workers) and a national survey of 12-19 year old adolescents, all of which were conducted in each of the four countries.
Sexual Coercion at Sexual Debut in sub-Saharan Africa
Gender scripts play a significant role in establishing the way sexual interaction takes place.19 Previous research in sub-Saharan Africa has concluded that gender roles in parts of the region sexually disenfranchise girls and make them more vulnerable to sexual assault.20-26 Research done on adolescents' gendered socialization in Africa has found that adolescent relationship dynamics are typically characterized by gender inequality that gives preferential treatment to male decision-making and that there is poor couple communi-cation about sexual matters.20;27;28
One result of this gender inequality is that it is not commonly acceptable for girls to be sexual agents or demonstrate sexual interest. Boys' awareness of this coyness teaches them to not respect a woman's `no.' As an urban boy in South Africa related, "When [women] say "no" they mean "yes." [A woman] can never come out clearly and say "Let's do it." You need to read her facial expression If she keeps on saying "no" and closes her eyes, she wants it [sex]."29, pp. 163-164 These gender roles for women are coupled with sexual scripts for men about what it means to be a real man (not taking "no" for an answer) and male sexual voracity: Men are socialized to think that they cannot sexually control themselves.
Assumptions about women's sexuality condone sexual violence. Research in South Africa found that girls' exhibition of sexual unwillingness, which maintains her respectability, opens the door to socially acceptable coercive sexual behavior including violence.29 In a study conducted in Nairobi, Kenya, young people stated that what the respondents called "soft rape" (penetration occurring after the girl has sexually "teased" a boy) was acceptable.30 Kalichman et al. (2005) found in their study of men and women in Cape Town, South Africa that 27 percent of participants across genders agreed that rape is usually a result of something a woman says or does, 18 percent felt that some cases of rape involve a woman who wants to have sex and 29 percent said that rape is often a woman's fault.20 Ideas such as these make it hard for women to convincing demonstrate to their partner that they do not want to have sex.
Inadequate social and legal sanctions create an environment in which sexual coercion happens largely with impunity. Perpetrating sexual coercion is not a highly stigmatized behavior and prosecuting a perpetrator remains extremely difficult, as evidenced by the Kobe Bryant rape trial of 2004 in the United States.31 In some countries, family members may take matters into their own hands and perpetrate revenge upon men who force sex on their wives and daughters. Even in countries with laws, there are significant barriers for females to report these negative and potentially stigmatizing experiences.
Focus group discussions with 14-19 year olds in Burkina Faso, Ghana, Malawi and Uganda that were conducted as part of the same study as the results being presented in this paper found that rape or forced sexual intercourse came up spontaneously in the context of a child being forced to have sex (being "defiled") or in the context of what some men, after drinking alcohol or smoking hemp or marijuana, will do to young women. The latter context was related to men's "uncontrollable" urges or willingness to engage in unprotected sex when under the influence of alcohol or drugs.32 One common theme in the discussions from Burkina Faso was men drugging young women in order to have sex with them. In some focus groups, rape or forced sex were described as a response to young women refusing sex, even after attempts by young men to negotiate for sex, particularly if young women received money or gifts (mentioned among focus groups in Malawi). Sexual coercion within marriage or in boyfriend-girlfriend relationships was rarely mentioned.
Measuring the prevalence of coercion is difficult because there are many different ways that coercion can be conceptualized. Furthermore, reports of coercion are subject to personal interpretation based on perceptions of entitlement and gender roles in addition to recall bias. Although there has been growing interest in research on sexual coercion in sub-Saharan African2;6;8;30;33-40, most studies in the region have focused on the prevalence of non-consensual sex among females in general. Very few studies have looked at the sexual coercion experiences of adolescent females in sub-Saharan Africa at their sexual debut. Studies conducted in South Africa, Kenya, Ghana and Uganda have found that the proportion of adolescent females who reported that their first sexual encounter was coerced ranged from 14 percent among a rural population in Uganda to 32 percent among urban residents of Cape Town.18;41-47
There have been a number of qualitative studies in sub-Saharan Africa that have shown that young people have sex for economic reasons. Items such as food, gifts, clothing, books and toiletries have been identified as encouragements for young women to have sex.8;32;37;38;40;48-52 While the giving of gifts cannot uniformly be interpreted as a coercive force to have sex, even the giving of luxury goods may influence girls to have sex as payback or perceived obligation. The prevalence of this form of coercion is not limited to taking place between a male and his sexual partner. Parents also pressure their daughters to have sex for economic reasons.
It is clear to see that in spite of the recent attention to the issue of sexual coercion, what have been lacking are nationally representative studies on the subject. The results presented in this paper provide those numbers for the four countries included in the study. The qualitative evidence shed further insight into the trends that the quantitative data represent.
Data for the study are derived from national surveys and in-depth interviews (IDIs)all conducted with adolescents.
Nationally Representative Survey
Nationally-representative household surveys1 on adolescent sexual and reproductive health were conducted in 2004 among 12-19 year old males and females: 5,950 in Burkina Faso, 4,252 in Ghana, 4,012 in Malawi, and 5,065 in Uganda. The sample, which covered all 12-19 year-old de facto residents in private households, was a two-stage stratified sample design: district and households. When there was more than one eligible 12-19 year old in the household who was interviewed, one of the respondents was randomly selected to answer an additional section with questions on sensitive topics including physical and sexual abuse. The survey was pretested, modified accordingly, and then translated into the major languages spoken in each of the four countries. The translation was pretested again before the instrument was finalized. On average, each interview lasted approximately 55 minutes.
Training of the field personnel was based on standard Demographic and Health Survey (DHS) training protocols for conducting an interview. Further DHS protocols were followed regarding making callbacks and completing survey questionnaires. Same-sex interviews were conducted because of the sensitive nature of the topics covered. Consent was obtained from all young people before they participated and parental or guardian consent was also obtained for people younger than 18 years.
Ensuring privacy of the interview was absolutely critical to fielding the survey, and interviewers were trained to conduct interviews in places or ways that would assure privacy for adolescent respondents. The section on physical and sexual abuse was not to be administered if anyone older than 3 years was within hearing distance.
Approximately 55 in-depth interviews were conducted with females ages 12-19 in each of the four study countries and consisted of in- and out-of-school adolescents recruited from urban and rural locations (Table 1). In addition, interviews were conducted among young people in specific groups that were considered to be at higher than average risk: young married women, women with children, refugees (Ghana and Uganda) and petty traders. The interviewers were the same sex as the respondent, they took place in a neutral location, and they lasted between 30 minutes and 2.5 hours. The discussions were tape- recorded, transcribed and translated from local languages into English and, in the case of Burkina Faso, into French.
Methods of Analysis
For the quantitative data, the dependent variable of interest is the variable on willingness at first sexual intercourse: "Thinking about the first time you had sexual intercourse, would you say you were very willing, somewhat willing or not willing at all?" The response categories are "very willing," "somewhat willing" and "not willing at all." Sexual coercion is being defined as having answered "not willing at all." The independent variables being included in the paper are country of residence (Burkina Faso, Ghana, Malawi and Uganda), place of residence (urban/rural), age at first sexual intercourse (<12 years of age, 12-14 years of age, and 15-19 years of age), schooling status at time of first sex (in-school/not in school); received sex education (never, before first intercourse, after first intercourse); relationship with first sex partner (boyfriend, husband, live-in partner, and casual acquaintance/other); age difference with first sex partner (partner is 10+ years older than respondent, partner is 5-9 years older, partner is 1-4 years older, partner is the same age or younger than the respondent, or respondent doesn't know partner's age), orphanhood status (both parents alive; mother died, father alive; father died, mother alive; both parents died) and wealth status (quintiles). The wealth status was determined using Filmer and Pritchett's (1998) method which uses principal component analysis of information on whether or not a household had amenities and assets such as piped water, electricity, car, TV, etc.53 Wealth quintiles were determined separately for urban and rural households before combining the sample. The quantitative analysis of the data was done using STATA 8.0 (StataCorp, College Station, Texas).
The IDI transcripts were coded using N6 qualitative software (QSR International, Doncaster, Australia). Each individual who participated in an in-depth interview was treated as a unit of analysis. All coercive experiences at sexual debut were identified and summary matrices were used to write summary text was which was compared by at least two members of the study team with the summary matrices to ensure that any one analyst's subjective biases did not determine the conclusions. Text was written based on common themes arising from the summary text.
A word must be said about interpreting unwanted sexual debuts from the in-depth interview narratives. In a few of the situations where the girl identified her sexual debut as forced, the emotions expressed surrounding sexual debut do not match the description of the sex as coercive. In Malawi, the case of an urban, out of school 18 year old demonstrates how this occurred:
I: How did this come to happen?
R: He forced me.
I: How did it all start?
R: I have forgotten.
She goes on to describe a non-forced situation: "He invited me to his place. While there he asked if I could spend a night there. At first I refused but I gave in after his sister supported the idea, and this was my first time to have sexual intercourse with him." She said she felt "natural, we are supported to do it." She said that while at first she was frustrated because she did not know how it was done, later on she was happy with it and now when she remembers, she's just surprised how it all happened. As the narrative in toto does not describe a negative or unwanted experience, for the purposes of this analysis, narratives that included elements such as this one were not considered coerced. Their interpretation is taken up again in the discussion.
The percentage who were "not willing at all" at sexual debut was the lowest for Burkina Faso (15 percent), followed by Uganda at 23 percent, Ghana at 30 percent and Malawi at 38 percent. Turning briefly to the other response categories, between 20 and 30 percent were "somewhat willing" in all four countries. Since it is less clear what this response means, for the purposes of this analysis, these respondents are not categorized as having experienced a coercive sexual debut. This leaves less than half of the females in Ghana, Malawi and Uganda indicating that they were "very willing" at first sexual intercourse. Burkina Faso is the exception, but does not depart radically from the trend. Just over half of the females in Burkina Faso were "very willing" (Table 2). The rest of the analysis presents results for those who reported that they were "not willing at all" at sexual debut.
When narrating the context of first sex experiences in the IDIs, unwanted sex emerged as a frequent theme of females' debuts. The most prevalent forms of coercion across all four countries were force; pressure from money or gifts; male's flattery, pestering, and threatening to have sex with other girls; and girls' passive acceptance.
Forced sexual intercourse
In the IDIs, experiences forced sexual debuts emerged as a frequent theme. Some females in all the four countries reported being forced to have sexual intercourse at sexual debut. In virtually all cases, the victim knew the perpetrator who was either a boyfriend or a casual acquaintance.
Interviewer: And what happened?
Respondent: He started fondling my breast and I asked him why he was doing that. He told me he was just playing with me. Then all of a sudden he pulled me to the bed and had sexual intercourse with me.
I: After the sexual intercourse who did you report to?
R: I did not tell anybody because my father had already warned me about them [the teachers in training, one of whom raped her] and had told me not to fetch water for them again. So I was afraid to tell my parents.
I: Did he [the rapist] warn you not to tell anybody?
R: No, he did not. But I was afraid that if I told my parents my father will beat me.
Urban, street child, 14 years old, Ghana
One of the themes that emerged in the narratives of adolescents who experienced forced sex at sexual debut was the fact that they were trapped by the male with whom they were either in a romantic relationship or he was otherwise known to female.
I told him that I did not seek permission to sleep out from my parents for when I was leaving home I just told them that I was visiting a friend but he didn't take any of that and just locked me in and gave the keys to one of his friends through the window only to return with them the following morning. So during the time we were there I had sex with him.
Urban, 18 year old, Malawi
One day I was waiting for a friend of mine and was seated along the path. He grabbed my shoes and took them. He told me I will never see them if I don't go to his place. I feared to go home without the shoes. So I followed him When I reached in his house, he closed the door and forced me into sex. I could not scream because I was in his house and dreaded the embarrassment it would cause I felt bad and regretted why I had gone there but I did not tell anyone. I kept it a secret. It was very painful. I cried a lot but dried the tears. I went back home and pretended as if nothing happened.
Rural, 16 year old, Uganda
He deceived me to have sex with him and that is how our relationship started [She laughs] You know, that boy forced me into having sex, he played about with my head. He told me to go to his home for exam papers, by then we had started our exams. So I went. He forced me into sex. I tried to fight but he over-powered me. He kissed [me] then he had sex with me. [She shakes her head.] I felt so bad, but I forgot all about it.
Rural, 19 year old, Uganda
There were also situations where friends/peers of the girl facilitated the unwanted sex:
Anyway, it was like this. My friend is the one who took me to his house. She told me that, "You escort me," and we went to the house of her friend. When we were there we saw that boy also coming. Then she told me, "You be talking with that boy; I am also talking with this one outside here and there is nothing he will do to you." I also started talking with him knowing that she was outside. But she had gone and the boy forced me into sex.
Rural, 19 year old, Uganda
Some girls reported resorting to evading the males to avoid unwanted sexual intercourse. According to a rural, Malawian 19-year old, she had been evading her boyfriend with excuses until he cornered her: "Then he said to me, `I have come for that issue [sex].' Then I had no choice but to do it." This relationship wound up in marriage and the respondent related how now she is forced to have sex with this partner virtually everyday.
In two of the cases from Malawi where forced sex was reported, the girls exhibited a notable amount of agency after the sex. An urban, 18 year old said she wished that he had asked her before bringing her to the lodge where he forced her to have sex and an urban, 18 year old let the man who forced her to have sex know he had done something wrong and because of that experience, she ended the relationship with him. Decisive, clear communication from the girl such as these examples cited above was anomalous.
Pressure from Gifts or Money
The role money or gifts play in leading to sexual intercourse warrants analysis in its own right and this has been done elsewhere.54 As was demonstrated in that analysis, in the majority of the narrations of sex involving money or gifts, the respondent did not make clear the role that money played in bringing about sexual intercourse. In Malawi, gifts were generally not described as having a coercive influence. In comparison, the vast majority of the sex associated with money or gifts in Uganda was described as coercive. Its treatment here is limited to when the transactional interaction becomes coercive: when receiving money or gifts compromised the girl's ability to say "no."
A Burkinabé female described how she was lured by money/gifts to put herself in a situation where sexual coercion occurred. The boy in the narrative below had invited the respondent to go to a video club, an informal movie viewing location, to watch a movie:
R: He charmed me and then we had sex. It was my first time. [ ] He encouraged me to go with him to the bedroom so that he could give me a gift; when we went into the bedroom, he shut the door.
I: But when he shut the door, did you try to scream?
R: No. I wanted to scream and he told me not to cry and I shut up.
Urban, 19 year old, Burkina Faso
A Ugandan, rural 15 year old girl related, "The man told me that `If I have sex with you I will use a condom and nothing will happen to you, I will also give you 2000 shillings' so I accepted because I needed the money."
Sugar daddy relationships were not the norm among the respondents. The following Ugandan who was in a relationship with a man 15 years older than her was the only one to narrate a clear sugar daddy relationship.
He would pick me from home secretly and take me for film shows in town. I would always lie to my mother that I had gone to my Auntie's place and would spend nights with him.
At the end of it all he asked me to show him that I loved him by having sex with him and I complied. I could not refuse because I was ashamed of all the things he had done for me.
Rural, 17 year old, Uganda
This relationship resulted in a pregnancy and only after she was pregnant did she find out he was married and had children with another woman.
Men's flattery, pestering, and threatening to have sex with other girls
Another way that unwanted sex took place was through the male's use of flattery or otherwise playing on the emotions of the girl to have sex against her will. The way the situations were described, the circumstances placed girls in compromised positions that hindered their ability to say "no" to sex.
R: He came to find me here, we went to a small drinking place in the scrubland and afterwards we continued.
I: Did he force you?
R: No, he flattered me. I didn't want to accept because it was my first time and I didn't want to lose my virginity.
I: Afterwards what did you think?
R: I cried all day, I regretted it
Urban, 18-year-old, Burkina Faso
Out of a desire to please her partner, possibly to maintain the relationship, girls acquiesced to sexual intercourse under varying degrees of duress.
R: We talked about it and I told him we had to protect ourselves. And as it was my first time, I didn't want to, and he said it wasn't going to kill me. I still wasn't in agreement and he wanted to call his uncle to persuade me.
I: But you finally gave in?
R: Yes, I gave in.
Urban, 19 year old, Uganda
Girls in romantic relationships related being pestered by their partners to have sex, experiencing emotional isolation after turning them down. This girl's narrative highlights how her partner was angry with her for refusing to have sex which may have resulted in her engaging in unwanted sex shortly thereafter.
The first time he asked to have sex and I refused, we spent about a week when we were not talking. It was like our relationship had ended. But he would send his friend to tell me not to end the relationship. He was annoyed and angry with me. After that week, we went to his place and this time he forced himself on me and we had sex I was hurt, every part of my body was hurting. I got the flu. I was very disappointed. I did not even want to see anyone of them, the boy and his friend. That day he forced me, I abused him, and because of anger he slapped me twice.
Urban, 15 year old, Uganda
Having sex out of fear that if she did not give in to partner's demands he would have sex with other women was another motivation to engage in unwanted sex.
Girls' Passive Acceptance
Giving in to men's demands for sex occurred out of obligation, passivity and fear of the consequences of not giving in. Girls described this as "accepting what he wanted," protesting but giving in, and "letting him do what he wanted." Girls related how they had intercourse at the males' behest:
He told me what he wanted and I accepted.
Urban, 14 year old, Malawi
I: What did you do to show your disagreement?
R: I protested that I did not want but he insisted until what he wanted was satisfied.
I: Did he rape you?
R: No, he didn't rape me.
I: So you simply gave in?
R: Yes, I gave in I didn't feel anything because he just forced me.
Urban, 18 year old, Malawi
I: How did this come to happen?
R: We were in bed and he said we should have sex and I obliged.
Rural, 18 year old, Ghana
Sometimes girls allowed the unwanted sex to occur because to have cried for help would have led to being discovered with the boy. Feared consequences of being discovered with the boy included getting beaten by a parent and getting blamed for having put herself in a sexually vulnerable position.
At first I agreed [to have sex]. But after getting a second thought, I tried to refuse because I had never done it before. I was afraid of the outcome. I thought that I would fall sick after doing it, like getting a fever. But as I was thinking about this, my boyfriend was undressing me. This is when I realized that he was naked. He hid my clothes. I thought of yelling but this was my boyfriend and people would have asked why I had gone into his house. I went there knowing that he is not my brother and I knew that he wanted to have sex with me before even going there. But I was just kind of afraid.
Urban, 19 year old, Uganda
A Malawian, urban, 18- year- old girl who had just stepped out of the shower found her boyfriend hiding in her room. The boy took her towel off and she related:
He told me that...he couldn't let me get dressed. Then he told me that he wanted to have sex with me and he went on to say that he cannot control himself. Then I was dumbfounded because to scream would create a scene so I just let him do what he wanted not realizing that I will end up being pregnant.
This respondent had told him to wait until they were married and "he said that he cannot do that" because he had seen her naked (and therefore could no longer control himself). She denied herself the assistance of family members who were present who could have intervened because she was convinced that the trade-off was not worth it. The way she put it was that it "would create a scene."
Another manifestation of passive acceptance is treating coercion as normal, as some of the respondents did. The guiding assumption in these cases is that all men coerce. This is one of the dimensions of gender socialization that tends to perpetrate the acceptance of coercion.
[After having experienced forced sex,] he started sending my friend, a girl, because he knew I was mad at him and did not want to see him again. My friend convinced me that such things happen to every girl so I should get used [to it]. So I forgave the boy and went back.
Urban, 15 year old, Uganda
I: Is it possible to refuse having sex?
R: You cannot refuse because my sisters have told me that whatever you say, he will find a way of getting you to accept. He might romance you until you get the feeling as well. [ ] If we are somewhere with the boyfriend, I would move away because if you refuse he will not feel good...I am not sure what else I would do to get away from him if he insisted on having sex with him. You might find yourself having sex in case you do not want to embarrass him. Boys turn it into war if the girl refuses to have sex with him.
Urban, 17 year old, Uganda
When coercive sex becomes so normalized that it is treated as inevitable, girls are taught that their sexual interests are irrelevanta fundamen-tally disempowering and therefore dangerous perspective to have as the locus of sexual impor-tance rests solely with the male.
While these narratives allow for a deeper understanding of how unwanted sex occurs, they are not able speak to who is most at risk. We turn back to the quantitative data to shed insight on how risk is distributed in the population.
The next section analyzes some of the factors that have been found to be associated with a coercive sexual debut (Table 3). Chi-squares were conducted to determine which variables where significantly associated with willingness at sexual debut on a country by country basis.
Place of residence
There is no definite pattern by rural/urban residence. Ghana was the only country that demonstrated a significant difference in the probability of experiencing a coerced debut based on residence. In Ghana, residing in a rural area is protective of experiencing sexual coercion at debut.
Age at first intercourse
In Burkina Faso, Ghana and Malawi, the legal age of sexual consent for girls is 16 and in Uganda it is 18. In each of these countries, when the girl is a virgin and a minor, irrespective of whether it was consensual, sexual intercourse is deemed defilement and can result in prosecution of the male. While the law on defilement is rarely enforced, there is a great deal of fear among men of being charged with defilement when a pregnancy occurs.
Age did not come out as significant in Uganda as it did in the other three countries. In Burkina Faso and Ghana, younger age at sexual debut is correlated with being "not willing at all". While the cell sizes are very small for girls who had their sexual debuts before age 12 in Burkina Faso and Ghana, over half of these girls were "not willing at all." In Malawi, a different pattern emerges. The 15-19 year olds are the most likely to have been "not willing at all" at sexual debut compared to the other age groups: 32 percent of girls less than 12 at debut were "not willing at all" compared to 42 percent between the ages of 15-19 at debut.
Schooling status was significant in Ghana and Malawi, but not as one would expect: being in school in both of those countries increased the probability of experiencing an unwanted debut. Therefore, schooling is not protective of having an unwanted debut. Yet there could be coliniarity between schooling status and wantedness of sex: The qualitative data showed that girls in school are more likely to want to avoid sexual intercourse to prevent pregnancy as pregnancy indefinitely suspends and potentially ends their schooling.
Received sex education
Receiving sex education before becoming sexually active could equip young people with information to make better informed decisions in the sexual sphere and consequently increase their chances of avoiding unwanted sexual intercourse. This variable was only significant in Uganda where fewer adolescents who did not have sex education at sexual debut were "not willing at all" compared with adolescents who had sex education after sexual debut to have been "not willing at all" compared with adolescent who had received sex education prior to sexual debut or who had never received sex education were less likely that having received sex education would increase the probability of the adolescent identifying the experience as unwilling as most of the curricula in use in these countries do not cover the topic of consent.
Relationship with first sex partner
Type of partner was significantly associated with willingness across all four countries. The most common first sexual partner in each of the countries was a boyfriend. Results from Ghana and Malawi suggest that generally twice as many females aged 15-19 years who were not married indicated unwillingness to have first sex compared with females who were married. The exception was Burkina Faso where regardless of union status, roughly the same percentage of females (15 percent) stated their unwillingness to have first sex. The proportion "not willing at all" with a live-in partner looks similar to the proportion "not willing at all" with a husband in Ghana and Uganda, while in Malawi the proportion "not willing at all" with a live-in partner looks more similar to the proportion "not willing at all" with a boyfriend. While first sex with a casual acquaintance makes up a small proportion of first sex partners, sex at debut was most likely to be unwanted when the partner is a casual acquaintance.
As with schooling status, the results on this variable may be influenced by coliniarity. Girls whose first partners were their husbands might be less willing to say that their first sex was "not willing at all" because sex is expected within marriage and husbands are not commonly identified as perpetrators of unwanted sexual intercourse. Therefore, what is noteworthy is how high the proportion is of girls who said that their first sex experience with their husband was "not willing at all."
Age difference with first sex partner
Age difference with one's first sex partner is a proxy for power in the relationship. Only in Ghana and Malawi did the age difference with the partner emerge as significant yet the variables do not react the way we would expect. In both countries, having a partner 1-4 years older is more likely to be correlated with coercive first sex than having a partner 5+ years older. In neither country is the proportion having coerced first sex with a partner 10+ years older as high as it is for having a partner 1-4 years older or having a partner the same age or younger.
Orphanhood and Household Wealth
Orphanhood, defined as having lost either one or both parents, is a proxy for vulnerability. In Ghana, orphaned females are more likely to experience coercive sex than those who still have both parents alive.
Household wealth was only significantly associated with coercive first sex in Burkina Faso. The relationship functioned in the way we would expect: poorer girls are more likely to have experienced unwanted sexual debuts.
We used logistic regression to estimate the odds for being "not willing at all" at sexual debut within the study population (Table 4). For the purposes of this paper, we define our dichotomous coercive first sex outcome variable as follows: 1 = not all at willing at first sexual intercourse; 0 = willing/somewhat willing to have sex at first intercourse. Coefficients are expressed as odds ratios relative to the omitted reference category (OR = 1.00).
Age at time of interview is marginally significant in Ghana. Younger girls (12-14 years old) have a higher probability of having been not willing at all compared to older girls (15-19 years old). Living in a rural area in Burkina Faso makes one marginally significantly more likely to have experienced coerced sex. Age at first intercourse is not significant across any of the countries. As suggested in the bivariate table, schooling status at first sex does not behave the way one would expect. It is only marginally significant Malawigirls who are in school are marginally more likely to have experienced a coerced debut. Age difference with the partner is marginally significant in Malawi and Uganda but in opposite directions. Having a partner closer to you in age increases the probability of coercion in Malawi while it decreases the probability of coercion in Uganda.
Having first sex with a casual acquaintance is associated with significantly higher odds of being coerced across all four countries. Girls in Ghana were six times more likely to be coerced when their first sex partner was a casual acquaintance (OR = 6.00, p<0.001 significance level) than when the partner was a boyfriend compared to an approximately three and a half-fold increase in Uganda (OR = 3.53, p<0.001 significance level). In Malawi, those whose first sexual intercourse was with their husband were significantly less likely to report that they were coerced, possibly for reasons provided in the bivariate relationships section. Yet what is striking is that for the other three countries in this model, there is in fact no protective effect of being married: Girls are just as likely to experience an unwanted debut with a husband as with a boyfriend. In Ghana, in comparison to girls who still had two parents alive, girls who had lost either parent were at twice the risk of experiencing a coerced sexual debut. Burkina Faso is the only country where wealth has a protective effect: The wealthiest respondents are marginally less likely to have been "not willing at all" at sexual debut. In Ghana, the relationship between wealth and unwanted sexual debut is harder to interpret.
In spite of including in the model a set of variables that have been found by other researchers to be related to coercion at sexual debut across a wide variety of cultural settings, the R-squared values for the models are very poor. This demonstrates that sexual coercion is happening to young women irrespective of a host of characteristics that we would predict to be protective: age at sexual debut, schooling status, age difference with the partner, relasionship with first sex partner, orphanhood status and wealth status.
In three of the four countries in the study, less than half of the females were "very willing" at first sexual intercourse. The exception was Burkina Faso where the percentage "very willing" was 57 percent. This can be understood in part by looking at the relationship with first sex partner. In Burkina Faso, husband/live-in partner makes up a much larger proportion of first sex partners than in the other three countries and coercion is less reported within marriage. This is not to say that the sex is necessarily more wanted within marriage but rather there may be greater barriers to reporting the sex as unwanted within a formal union so that unwanted sex with a husband may be more underreported than unwanted sex with less formal partners.
The multivariate logistic regression showed that place of residence seems to work differently across the four countries with rural residence increasing the probability of being "not willing at all" in Burkina Faso. Contrary to what one would expect, age at first sex was not predictive of the sex having been "not willing at all." Being in school increased girls' risk of experiencing a coerced debut only in Malawi.
Having a partner closer in age was protective except for in Malawi where being close in age to one's partner in fact increases the probability of one's sexual debut being "not willing at all." But this is significant only at the 0.1 level. Relationship to one's first sex partner shows the most significant findings in this model: Sexual debut occurring with a casual acquaintance/other greatly increases the odds of the girl's sexual debut being coerced. Having one's sexual debut with one's husband significantly decreases the odds of being coerced in Malawi. Orphanhood status in Ghana may increase girls' risk of sexual coercion because these girls may be more socially and economically vulnerable. Another reason for their greater vulnerability is that they may be living with non-family members, giving males in that household greater sexual access to them. A third possibility is they may not experience parental monitoring to the same extent as non-orphaned adolescents.55
The qualitative data provide insight into how unwanted sex takes place: through force, the giving of money or gifts, flattery, wearing the female down or threatening have sex with other girls, and passive acceptance. Transactional sex was more prevalent in the interviews from Uganda than in Malawi. The consequences of being unwilling at first sex as related by respondents included severe body pain and getting sick. Some of the respondents had not told anyone about the unwanted sex experience primarily because they feared their parents' reaction to revealing that they had engaged in sexual intercourse. These fears serve to further isolate girls from social support that has the potential of helping them cope with the negative experience that they had. It also denies them the ability to take action against the boy, if they so chose.
As discussed in the methods section, in Malawi there were examples in the data of girls naming their debut as coerced but then going on to describe neutral or even positive emotions around the experience. This suggests the possibility that in the quantitative data the number of Malawian females naming their experiences "not willing at all" may be higher than what one would deem as coercion if more of the sexual narrative had been gathered. Kaler has discussed how in traditional Malawian languages, the use of the translated term rape can encompass a broader range of behaviors beyond forced sex that also includes aggressive male sexuality. This is perhaps why the number of girls who said their debut was "not willing at all" was the highest in Malawi.
A further complication in interpretating the results that impacts not only the qualitative data but the quantitative data as well is that within these societies that maintain traditional gender roles there remains the prohibition against girls being sexual actors. That includes girls showing sexual interest or initiating sex, especially as virgins. Consequently, it is socially desirable to abnegate responsibility for sex having occurred and one way to do that is to say it was forced or unwanted. While that may be influencing some respondents to relate their first experience as not what they wanted, the social prescript is in effect across the board in each of these countries and nevertheless, a significant proportion of respondents said that they were very willing at sexual debut. If the social prohibition against girls wanting to have sex were that complete, one would expect there not to be any girls who were very willing at sexual debut. Therefore, the fact that a significant proportion of girls are able to claim sexual willingness demonstrates that there is room for girls to disclose that they wanted to have sex and in fact lends credence to the unwantedness of the other girls' experiences.
Youth are in the process of developing as well as firming up their perceptions and attitudes about gender as they enter the sexual sphere. The early sexual experiences they have inform future sexual behavior which is why experiences at sexual debut remain an important area of study. What the study has done is to bring into the open some of females' experiences which are rarely discussed. It is important to address sexual coercion because it is one of the many forms of violence against young women: Sexual coercion has been found to be linked with sexual harassment and increasingly violent sexual behavior which reflect underlying structural factors that contribute to the oppression and exploitation of women.5;56-59 Authors of research conducted in Khayelitsha, Cape Town with adolescents conclude that "Violence was not limited to the first sexual act or to the first relationship, but was also reported to be a feature of all subsequent sexual relationships."6 To the extent that sexual coercion including rape is just the most extreme manifestation of a pattern of gender oppression that includes sexual harassment and gender-based violence, the prevalence and experiences of unwanted sex at sexual debut can be heeded as a measurable manifestation of more widespread abuse and gender oppression.
In the countries in which this study was conducted, it is not socially acceptable for girls to be sexually interested or initiate the sexual interaction. Furthermore, negating boys' and men's sexual advances and fighting off their sexual overtures is the only acceptable female response. This limits girls' ability to demonstrate their sexual interest, thereby making it difficult for men to distinguish between a willing partner and an unwilling partner. Poor sexual communication can pave the way for unwanted sexual experiences.
With an eye toward improving sexual health outcomes, understanding the sexual and reproductive behaviors of young people, especially young women, and the factors that protect or put them at risk of HIV infection, other STIs and unwanted pregnancy is critical. Youth ages 15-24 are one-fifth of the population of sub-Saharan Africa and their state of health has significant implications for the future of the individual countries included in this study and the region as a whole.60 Protective behaviors shaped through appropriate information and services could provide immediate and long-term benefits to youth and consequently society.
One way to address youth's sexual health is through providing sexuality education that specifically addresses consent and coercion. Traditional HIV/AIDS programs which emphasize abstinence, faithfulness in relationships, and condom use do not protect young women from HIV in unwanted sexual situations as condoms are rarely used when the sex is coerced.61 Unfortunately, sexuality education for youth rarely discuss sexual coercion.62 Therefore, rather than limit sex education to promoting abstinence and if the adolescents are lucky, partner reduction and condom use, as others have identified, other topics that need to be included are preventing sexual coercion, providing support to victims who have experienced unwanted and most likely unprotected sexual intercourse, strengthening the legal and advocacy environment for young women so that they feel like they have recourses against those who may coerce them into unwanted sexual activity, training providers and educating young women about their rights including addressing gender norms and teaching communication and negotiation skills.62 Police could play a role in strengthening girls' and women's ability to take action against perpetrators, the media could educate young people about girls' sexual rights and the importance of sexual communication, and community-based organiza-tions can help provide safe, socially supported spaces for young people to interact.62
There is increasing support for a response to sexual abuse in Africa as evidenced by the widely endorsed African Charter on Human and People's Rights and the Rights of Women in Africa. This is happening in conjunction with some African countries passing their own laws to punish perpetrators of violence against women. For example, in 2006 Ghana passed a domestic violence law to punish perpetrators of sexual coercion. Yet this of course will only have an impact in cases where the coercion is reported.
Much of what we know about adolescents' intimate relationships, especially in developing countries, is limited to a few indicators and frequently does not include the wantedness of the interaction. This paper adds to existing knowledge by providing additional information about girls' feelings about their sexual relation-ships, specifically focusing on their experiences of sexual coercion. Girls' experiences of unwanted sexual contact which place them at risk for HIV, other STIs and unwanted pregnancy have not been addressed in policy solutions to these social problems. The new evidence presented herein on the extent of the phenomenon demands that coercion be considered when designing policies to reduce girls' risk to HIV, other STIs and unintended pregnancy.
It is likely that there was underreporting of sexual behaviour by girls in the study overall as sexual behaviour, particularly among unmarried girls, remains socially stigmatized where the study was conducted. Therefore, there is most likely a selection bias, overreporting the sexual experiences of those who are willing to report being sexually active. As with any study, relying on respondents' renditions of their own behaviour, the results must be interpreted with these cautions in mind.
The variable from the surveys that was used to measure coercion was a close-ended question which only allowed for the respondent to give one of three answers. Feelings around sexual behavior are complex, frequently conflicting, and subject to recall bias and other temporal influences such as other sexual experiences. To ask the respondents to summarize their feelings about the experience into these close-ended categories oversimplifies and obfuscates the range of emotions that the respondents may have about their sexual debuts. Furthermore, these cross-sectional data only reflect the respondents' thoughts at a specific point in time. Their feelings about their sexual debut may have been reported differently if they had been interviewed at another point in time.
The sexual narratives described and discussed in this report reflect those of the young women who participated in the IDIs conducted in each country. However, the qualitative findings reported here do not represent the views held by young women in general in Burkina Faso, Ghana, Malawi or Uganda. Those who were willing to participate may somehow have a different sexual history, than adolescents who refused to participate. Therefore, these results may not capture the range of adolescents' sexual debut experiences.
This paper analyzes coercion only from the perspective of females. Therefore, while males experience coercion as perpetrators and as victims, treating coercion among males is beyond the scope of this paper.
It is likely that there was underreporting of sexual behavior by girls in the study overall as sexual behavior, particularly among unmarried girls, remains socially stigmatized where the study was conducted. Therefore, there is most likely a selection bias, overrepresenting the sexual experiences of those who are willing to report being sexually active. As with any study relying on respondents' renditions of their own behavior, the results must be interpreted with these cautions in mind.
This is only a cross-sectional study so it cannot speak to change over time in the prevalence or context of sexual coercion of girls at sexual debut. Our hope is that these numbers will henceforth be able to serve as a baseline so that future research will benefit from this assessment of the situation in 2003/4 in the four countries included in the study.
1In Uganda, the sample is not nationally representative because four districts in the Northern region had to be dropped from the sample due to security concerns during fieldwork.
© Copyright 2007 - Women's Health and Action Research Centre
The following images related to this document are available:
Photo images[rh07035t3.jpg] [rh07035t4.jpg] [rh07035t2.jpg] [rh07035t1.jpg] [rh07035t4a.jpg]