The Influence of Messages From Referent Groups On Engagement in Sex Risk Behaviors Among Ugandan Youth

Purpose: Adolescents in sub-Saharan Africa are at increased vulnerability to sexual and reproductive health problems. Where adolescents gain information about sexual and reproductive health can influence engagement in risky sexual behavior. While family is regarded as one of the most important influences on adolescent sex behaviors, the cultural norms prevalent in Uganda may limit such influence. Instead, teachers, peers and the mass media are the preferred source of sexual health information among Ugandan adolescents. Utilizing the Theory of Planned Behavior (Ajzen, 1985), the purpose of this paper is to examine the influence of messages from different referent groups on sexual initiation among Ugandan youth. Methods: Secondary analyses were conducted on a dataset accessed through the Inter-university Consortium for Political and Social Research (ICPSR). Original data were collected in Uganda over a 6 month period in 2004; 6,659 eligible adolescents were interviewed. Inclusion criteria were children ages 12-19 years and de facto or de jure members of the household. There were 32 topic area included in the survey. For these analyses, data from questions in the following topic categories were used: behavior, sex, sex education, risk assessment, sex knowledge, and demographics. A two-step hierarchical multiple regressions were conducted to predict sexual initiation by demographics and sources where adolescents received pressure not to have sex. Results: Sample size for these analyses was 409. In the first model, age, sex and urban/rural place of residence were found to be significantly associated with age of first sex. This initial model was statistically significant (F [4, 403] =85.77, p <.05). Second model included variables related to pressure not to have sex (e.g. parents, peers, siblings and teachers) and expected negative outcomes of sex. The second model was also statistically significant (F [9, 398] =39.645, p < .001) and accounted for 47.3% of the variance. Findings revealed that pressure from friends to not to have sex was the only factor negatively associated with age of first sex. Implications: While parents have been found to be the most reliable sources of sex-related information, the cultural norms of the Ugandan society might prevent such interaction. In Uganda, peers were the only salient influences and unlike previous studies, peer influence was found to be protective, delaying age of first sex. This has important implications for sexual health interventions in Uganda. Targeting changes to peer norms (as opposed to increasing parent-teen sexual communication) may in fact result in greater delays in engagement in sex. Peer led and modeled interventions are recommended to increase efficacy of sexual health interventions in Uganda. In terms of social work practice, this study reiterates the need for more comprehensive programs that are more network oriented instead of just targeting the individual. Also, given that gender is significantly associated with age of first sex, interventions might need to be modeled differently for boys and girls.