The BARHS survey was conducted in Lobatse and Francistown and contains reproductive health information on 2,400 males and females aged 13-18. Qualitative information is obtained from 1995 focus group interviews on the causes and consequences of schoolgirl pregnancy in Botswana, conducted among male and female university students.
The first part of the analysis examines male and female patterns of adolescent sexual behavior among the Tswana. Dependent variables of interest include the percentage of male and female adolescents who are sexually experienced, and their number of regular and irregular sexual partners during the past year. The second part of the analysis describes adolescents’ perceptions of the advantages and risks of becoming sexually active, and the third part focuses on the sources of information affecting adolescents’ reproductive health attitudes. Finally, multivariate logistic regression analyses are used to examine the effect of schooling variables, household headship, and sources of reproductive health information on the likelihood that an adolescent is sexually experienced.
Focus group interviews are used to provide detailed insights into adolescents’ motivations for engaging in sexual relationships, and to illustrate the potential role of the parents and the educational system.
Findings
Patterns of Adolescent Sexual Behavior
Although there are growing concerns about the rapid increase in HIV infections, schoolgirl pregnancy, and the economic difficulties of unmarried parents and their children, little statistical data are available on adolescent sexual behavior. The BARHS data allow us to shed light on this issue. As expected the percentage of adolescents who have had sexual intercourse is low for the youngest age group, but increases rapidly with age. The results also show clearly that males tend to become sexually active at an earlier age than is the case for females. Comparison of males and females aged 15-16 shows that 41 percent of males in this age-group are sexually experienced as opposed to only 15 percent of females. By ages 17-18, this gender gap has virtually disappeared. The literature attributes early sexual activity at least in part to diminished parental control over the sexual behavior of unmarried adolescents. Focus group interview participants also argue that female headed households are associated with diminished control of teenage sexual behavior:
Kids will do anything they like, because the [single] mother won’t be able to exert control over them […] They meet people, they meet men from outside. And these men, they sweet-talk them. And like, at the end of the day they get pregnant (male Tswana student)
The results do not provide any support for these claims suggesting that children of female-headed household exhibit less responsible sexual behavior, after controlling for age.
Interestingly, the results also show that education affects males and females differently. For females, the level of education attained has no clear impact on the likelihood of being sexually experienced, but being enrolled in school does. The percentage of female who are sexually experienced is much lower for those females who are enrolled in school than for those who are not (11 versus 27 percent for ages 15-16 and 53 versus 75 percent for ages 17-18). This effect may be caused by the combination of a relatively small number of pregnancy-related school dropouts, female boarding schools, and deliberate efforts to avoid sexual relations while in school. For males, the opposite pattern emerges. That is, being in school does not have a consistent effect on sexual experience, but level of education does. Male adolescent with secondary education are much more likely to be sexually experienced that those with lower levels of education, even after breakdown by age. The latter finding is consistent with arguments that males from higher socio-economic groups find it easy to attract girls. In addition to having a high educational status, these young males are likely to be relatively well-off, since students who pursue a secondary education are more likely than those who do not go to secondary school to come from families with a higher socio-economic status.
Consistent with the findings, the results confirm that older adolescents are more likely than younger adolescents to be sexually active, and that males become sexually at an earlier age than is the case for females. It further shows that it is fairly common for males and females alike to have one or more irregular sexual partners. For example, among females aged 17-18 22 percent report having one irregular partner, and an additional 16 percent indicated having two or more irregular sexual partners. For males, the corresponding figures are 26 percent and 21 percent respectively.
Perceived Advantages and Risks of Becoming Sexually Active
The above findings demonstrate that the traditional norms prescribing premarital abstinence are no longer adhered to, and suggest that different values operate in contemporary Tswana society. For females, the most striking finding is the fact that roughly a third of females believe that becoming sexually active has no advantages. Moreover, very few females mentioned any specific advantages of sexual activity. Marriage and childbearing, which were such important status indicators in traditional society, are no longer valued. For males, pleasure and status clearly stand out as the most important reasons for becoming sexually active (21 and 18 percent respectively for ages 17-18). Contrary to females, the percentage of Tswana males who think there are no advantages to becoming sexually active declines rapidly with age.
The low percentages of respondents who report any advantages of becoming sexually active appears inconsistent with the relatively high level of adolescent sexual activity in Tswana society. This suggests that there may be other motivations for becoming sexually active that are not adequately picked up by the survey questions. In the focus group interviews, respondents strongly emphasize that female (but not necessarily male) sexual activity is often motivated by economic factors:
We are living at times when, you know, when you are teenagers you have to dress, you have to get something, you have to be¼ You know, there is sort of a standard you have to live up to […]. And you know parents cannot afford to give their kids all this. And you know, it is like kids are forced somehow. They feel that they have to live up to that standard. That is why they end up meeting people and get pregnant. You would not find a teenager going out with a man who can’t give them anything (female Tswana student)
Nowadays, everyone wants money […] Young women, they want a guy who has money, who has a good car, who has a BMW or Mercedes-Benz, who can take care of them, who can provide for them (male Tswana student)
Teenage sexual activity may be discouraged by the knowledge of the potential negative consequences. It shows the percentage of Tswana adolescents who think becoming sexually active involves a risk because it may lead to stigma, pregnancy, or sexually transmitted diseases, including AIDS. Less than 10 percent of either males or females report that there are no risks associated with becoming sexual active. Female adolescents, even the youngest ones, are aware that sexual activity may lead to pregnancy. Among males, awareness of the risk of pregnancy becomes widespread a little bit later, mostly after age 15. Both males and females are well aware that sexual activity involves the risk of contracting sexually transmitted diseases. The findings strongly suggest that recent HIV prevention campaigns, and related campaigns, have been extremely successful in reaching and informing young adolescents about reproductive health risks, and that indicate that these efforts should be continued.
Sources Affecting Reproductive Health Attitudes
While these values represent a clear break with what was prescribed in traditional society, it is evident that these values are not (yet) consistent with contemporary patterns of sexual behavior among Tswana teenagers. It may therefore be helpful to examine from which sources adolescents obtain reproductive health information. Focus group interviews suggest that parents may not (or not yet) be able to successfully influence adolescent reproductive health behavior because parents typically do not discuss such issues with their children, but rather leave this up to the teachers:
It is not everybody in our society who can just sit down with their kids and talk about sex and related matters. But I think still that parents should learn to accept the situation as it is now […] Parents should face reality. Parents should begin to talk to their own children, if they can […] They think it is the duty of the teachers (Tswana female)
In most cases, we are not taught about reproduction at home. It is totally left to the teachers. The parents, I don’t know whether they are shy to talk about it (Tswana male)
It shows the percentage of adolescents who indicate their reproductive health attitudes were influenced by peers or siblings, parents, school, the media, or religion. Peers, siblings, and parents have the strongest influence on female reproductive health attitudes (with only minor differentials by age). At first instance, the large influence of peers and sibling on female reproductive health attitudes may appear surprising, but one should keep in mind that peer influence may also discourage female sexual activity. Schools have a substantial influence as well, but that influence diminishes with age, suggesting that the influence of schools diminishes when adolescents leave school. The influence of media and religion on female reproductive health attitudes is negligible.
Males are less likely than females to report being influenced by peers, siblings, and parents, but the influence of these source increases with age. Likewise, the percentage of males influenced by schools increases rapidly with age, from 24 percent for the age-group 13-14 to 55 percent for the age-group 17-18. Unlike females, the media has a quite important impact on male attitudes. The percentage of males admitting being influenced by the media ranges from 8 percent for ages 13-14 to 32 percent for ages 17-18.
Factors Affecting Adolescent Sexual Experience
We now examine the factors affecting the likelihood that an adolescent is sexually experienced, using logistic regression analyses. The first model examines the effect of schooling indicators and of household headship, after controlling for age and place of residence. As suggested earlier, for males having secondary level education increases the risk of being sexually experienced, but neither current school enrollment nor the type of household have a significant effect. The second model shows the effect of various influences on reproductive health attitudes, after controls. Males whose reproductive health attitudes are influenced by the schools, the media, or their peers, are more likely than others to be sexually experienced. Those who are influenced by their parents, on the other hand, are less likely to be sexually experienced. The third model introduces all of these variables simultaneously. The results presented in model 3 show that having secondary education more than doubles the odds that a male is sexually experienced, even after controlling for age and other factors, but being enrolled in school or being raised in a female headed household has no significant impact. Males who report being influenced by school and the media are significantly more likely to be sexually experienced, while those who are influenced by their parents are less likely to have had sexual intercourse. After controls, peers and siblings have no significant influence on the likelihood that a male is sexually experienced.
The results shows that having secondary education has no significant effect on the odds that a Tswana female is sexually experienced. However, unlike the case for males, females who are currently enrolled in school are significantly less likely than other females to be sexually experienced. As yet, none of the sources of reproductive health information have a significant effect on female sexual behavior. The finding that there are such striking differences in the factors affecting male and female adolescent sexual activity has important policy implication and merits further investigation.
