Haitian Women's Role in Sexual Decision-Making: The Gap Between AIDS Knowledge and Behavior Change

II. Presentation of Findings

1. Introduction to the Findings

This section draws from transcripts of focus group recordings to report findings on principal themes and sub-themes which emerged from the discussions. Highlighting areas of consensus and contradiction, we examine common beliefs, expectations, and assumptions behind much of the taken-for-granted behavior that is associated with the status of women in the sexual relationship and hence with the spread of AIDS. We also assess women's perceptions of their capacity to counsel others on protecting themselves from HIV.

In Section 2.1, we introduce the findings with a discussion of terminology and present characteristics of focus group participants by category. To gauge the participants' level of understanding of HIV, we summarize discussions of transmission from both the men's and the women's focus groups in Section 2.2 and comment on the similarity among groups. We then report findings from discussions among women participants concerning their perceptions of risk, their sense of personal vulnerability to the disease, and what they believe are the consequences of AIDS for families (Section 2.3). In Section 2.4, we report findings on prevention, including women participants' knowledge of, and concerns about, self-protection, as well as their views on the rights of women to intervene in sexual behavior that increases their risk of contracting HIV. The discussion highlights women's attitudes toward male sexual freedom and condom use in relation to female participants' perceptions of women's capacity to initiate behavior change. Section 2.5 then turns to men's perceptions of women's rights in the sexual relationship. This section reports comments from men's focus groups concerning traditional male expectations of female behavior and indications of a possible shift in attitudes as the AIDS epidemic advances. We focus particularly on men's views on how women can avoid contracting HIV, including strategies they think women should use to avoid sex with a promiscuous partner or to initiate the use of condoms, and how male participants believe men may react in such situations.

Section 2.6 deals with the way women see their role as advisors on AIDS prevention to two groups whom they might be in a position to influence: adolescent children and other women in their families and communities. It also includes women's views on the special vulnerability of young people and the range of responses they would expect their intervention to elicit from adolescents and their female peers.

Section 2.7 summarizes and discusses the findings presented in this report in relation to the study objectives outlined in Section 1.2.

A note on terminology

The investigators have attempted to distinguish focus group participants' views on women in relatively stable relationships ("maryaj", "plasaj") from their opinions on people in less stable sexual unions ("vivavek", "rinmin", "tizammi", "ménaj"). Moderators were instructed to use neutral terms such as "nèg" (guy), "fi" (girl or young woman), and "patné" to mean partner, whether temporary or well established. However, the participants, themselves, consistently used the terms, "mari" (husband) and "fanm" (wife), which made it appear that, for the most part, they were interpreting questions in the context of a stable relationship. The terms, husband and wife, did not occur in the story of Joujou and René, and the children were attributed only to Joujou, but the fact that they all lived under the same roof apparently led most respondents to define the couple as "maryé" or "plasé" and to assume that the children were René's responsibility. By interpreting the story and the moderators' questions in these terms, participants structured the problem according to cultural definitions of gender relationships.

As used by focus group participants, the terms husband and wife generally applied to any union in which a man and a woman are living together under customary rules that define a conjugal relationship. Participants usually referred to a wife as a woman who lives with a man, keeps house for him, provides sexual gratification, and bears his children. The terms "fanm", or "madanm", implied that, in exchange for material and emotional support, the wife, or woman at home, would not only carry out her domestic duties but would give up any sexual relationships she might formerly have had and remain faithful to the husband, or man who supported her. However, participants occasionally used these terms in the plural, distinguishing between the wife at home and other "wives" elsewhere. Since some men maintain more than one household, the plural reference to "wives" probably indicates long-term unions with more than one woman simultaneously, in contrast to more casual, transient, and uncommitted relationships.

While women would prefer that their steady partners also turn their backs on other women, fidelity seemed to be a luxury that few could expect. That is not to say that women in stable relationships never have outside men, themselves. Focus group participants of both genders acknowledged that "wives" as well as "husbands" may have "outside" partners who are responsible for the transmission of AIDS first to the women and ultimately to their husbands, or primary partners, as well. This report uses the term primary partner interchangeably with the terms husband and wife to include married ("maryé") as well as common-law ("plasé") relationships, two terms which are also used interchangeably. Although the taxonomy of sexual relationships in Haiti is differentiated by several types of union mentioned above, the authors use the term, "vivavek", as a generic adjective for unions which are relatively less stable and less binding than marriage, whether legal or common-law.

It is also worth noting that, while moderators asked their questions in the third person, participants often responded in the first or second person. To encourage free expression, the focus group guide phrased most questions in general terms that could be interpreted by participants as invitations to express an opinion or to share the experience of people like themselves. In much of the discussion, however, participants shifted the perspective away from the generalized other and toward a more personal syntax in which they obviously identified themselves (I/we) or their peers in the group (you) with the issues under discussion.

Segments of dialogue from the translated focus group transcriptions are reproduced throughout Section 2 to illustrate the way participants expressed their views on issues related to the study objectives. An identity code composed of the letter P and a number identify each speaker in the group. Comments which could not be so identified are designated PX.

Participant Characteristics

Appendix A contains a summary of focus group participant characteristics across the six categories as collected by local facilitators prior to the discussion sessions. There was little variation in mean age (29 years to 31 years) and a similar range in each category from 20 to 40 years (women) and 25 to 45 years (men). Savanne women in stable relationships tended to have more children (mean = 3.9) than Delmas women, and "vivavek" women and men in Savanne had the fewest (means = 1.9, 1.7).

Socioeconomic data confirmed the initial assumption that participants in this study would have little education and very low income levels. Income was consistently low, ranging from 54 percent of men in Savanne to 100 percent of non-factory women in Delmas reporting no stable income. This finding is not unexpected in view of the political and economic crisis a few months earlier which had resulted in the closing of many factories and workplaces. Among participants who reported income, the majority earned less than H$3 per day.

The chief difference between participants of Savanne and Delmas was in women's education levels. "Vivavek" and stable women in Savanne were particularly disadvantaged in this respect, with less than 25 percent literacy and school attendance. In contrast, sixty-seven percent of Delmas women factory workers and 76 percent of other Delmas women had attended at least primary school; 70 percent and 76 percent respectively said they were able to read and write. Men at both sites were more likely to have had formal schooling (Savanne 80 percent, Delmas 90 percent) and to be literate (Savanne 73 percent, Delmas 79 percent).

Among Delmas women, marital status proved to be a difficult item for the facilitators who completed the demographic data forms. Since recording errors invalidated this item for over half the female participants in Delmas, the distributions summarized in Appendix A do not include Delmas women. Although the data for men in Savanne and Delmas and for the female categories in Savanne were intact, they deserve cautious interpretation. The lack of mutual commitment in many unions is associated with frequent change and multiple relationships that may not have been reported. In addition, participants may not always have been willing to admit to local facilitators that they lived in unstable unions, and facilitators may have been uncomfortable about revealing information which they believed would be viewed negatively by the research team.

The demographic summary reveals low contraceptive rates, especially in Savanne, where none of the vivavek women and none of the male participants reported current use of a contraceptive method. Men in Delmas had the highest reported current rate, 61 percent, but 41 percent of the Delmas women factory workers and 52 percent of other Delmas women said they had never used family planning.

return to top of page