The use of focus groups to investigate sensitive topics: an example taken from research on adolescent girls' perceptions about sexual risks

O uso de grupos focais para investigar temas de natureza íntima: exemplo de uma pesquisa sobre pontos de vista de meninas adolescentes acerca dos riscos do sexo

ARTIGOS ARTICLES

Dora Lucia de Oliveira

Departamento de Enfermagem Materno-Infantil, Escola de Enfermagem, Universidade Federal do Rio Grande do Sul, Rua São Manoel, 963. Santa Cecília. 90620-110 Porto Alegre RS. [e-mail]

ABSTRACT

The methodology of focus groups has been increasingly employed in the context of research in the social sciences, particularly in health-related inquiries. Considerations about the sensitive aspects of such research are not, however, very often seen in research reports or discussion on ways of conducting sensitive research. The scope of this paper is to share an experience of conducting focus group research on sensitive topics, such as AIDS, risk and sexual issues, highlighting some methodological issues. More specifically, it suggests ways of working with teenage girls in focus groups about sensitive topics. The advantages of the use of focus groups to explore views on HIV/AIDS and other sexual risks are also discussed. Socio-cultural approaches to risk and feminist thinking permeate the main arguments.

Key words Focus groups, Qualitative research, Sexual behavior, Adolescent health, Risk

RESUMO

A metodologia de grupos focais tem sido cada vez mais usada em pesquisas no campo das ciências sociais, particularmente em investigações relacionadas à saúde. Considerações sobre os aspectos de natureza íntima que permeiam estas pesquisas não são, contudo, muito frequentemente vistas em relatórios de pesquisa, nem em discussões sobre as maneiras de se conduzir tais pesquisas. O objetivo deste artigo é compartilhar uma experiência de pesquisa com grupos focais sobre temas de natureza íntima, tais como AIDS, risco e sexualidade, destacando algumas questões metodológicas. Mais especificamente, o artigo sugere maneiras para desenvolver este tipo de pesquisa com meninas adolescentes. As vantagens do uso de grupos focais para explorar pontos de vista sobre HIV/AIDS e outros riscos sexuais são também discutidas. Abordagens sócio-culturais sobre risco e o pensamento feminista formam os principais argumentos.

Palavras-chave Grupos focais, Pesquisa qualitativa, Comportamento sexual, Saúde do adolescente, Risco

Introduction

As many have argued, AIDS is more than a disease, it is a social phenomenon. In order to expand our understanding of AIDS beyond its biomedical limits we need to investigate the social and cultural elements that make it a social experience. The study of people's ways of making sense of AIDS is perhaps the best means to carry out such an investigation, which includes the qualitative exploration of meanings, perceptions, attitudes, beliefs and experiences. AIDS social research frequently deals with the revealing of intimate data, such as personal information on drug use and sexual experiences. Considering the traditional taboos and prejudices involving sexual relations, one can argue that the specific research on the social and cultural aspects of HIV/AIDS sexual transmission constitutes a sensitive field of inquiry.

The contemporary increase of the number of feminine HIV/AIDS cases has put women in the limelight of AIDS social researches^1-5. Considerations about the sensitive aspects of such researches are not, however, very often seen in research reports, nor discussions about ways of conducting sensitive research.

The approach to sensitive research reported here is informed by feminist perspectives and was used in a focus group-based study about teenage girls' ways of seeing HIV/AIDS and other sexual risks. The aim of this paper is not to discuss the research results but to share the experience of doing sensitive research, highlighting some methodological issues. The emphasis on research methodology does not mean, however, to consider methodological issues as more important than the studied phenomenon. Qualitative research design is always dependent and secondary to the research object.

The research background

The research was designed to answer the question of 'how do teenage girls see the personal risk of catching HIV/AIDS in heterosexual relationships?; 'how' in terms of viewpoints, but also of the processes through which those viewpoints are constructed⁶. It was based on the assumption that if we are to understand HIV/AIDS sexual risk-taking and responses to the promotion of safe sex behavior it is crucial to map out the meanings attached to HIV/AIDS sexual risk-taking and the processes through which people make sense of it.

Also, it was assumed that the way people see sexual risks is not 'immune' to what they know about the world they live and its meanings do not stand on their own, for they are part of a much wider and general risk discourse that has been increasingly used in contemporary western societies. Such assumptions were based on sociological and anthropological theories of risk, like those developed by the sociologists Anthony Giddens⁷ and Ulrich Beck⁸ and the anthropologist Mary Douglas⁹. Those theorists have argued that 'risk' is an epistemological construct, that is, it exists only in terms of our knowledge. This argument did not mean to believe that the dangers of everyday life are not real threats, but that danger does not become risk until it comes to be known as such. The idea that 'risk' is an epistemological construct led to further considerations: first, that the 'process of knowing' in which risks become recognized as such involves judgmental considerations and second, that this 'process of knowing' is framed by the sociocultural environment where what comes to be known as 'risk' is to make sense.

The sociocultural nature of 'risk' indicates that risk meanings cannot be externally imposed, for they are dependent on what we value in life and also on the knowledge we use to make sense of it, personal values and acquired knowledge being context-dependent^9,10. We are not static and isolated entities and our beliefs are not separable from public discourses. We learn about the meanings of risk through social interactions and we act upon this knowledge in a social context¹⁰. To explore what health promoters' audiences understand by 'risky/safe sex' and the mechanisms through which this understanding is constructed is important for the theorizing of their responses to scientifically defined sexual risks, such as HIV/AIDS risk. In this sense, the research aim was to produce knowledge about the epistemological nature of sexual risk and HIV/AIDS risk and, also, to contribute to the sociological critique of the limitations of health promotion's definition of the problem of HIV/AIDS sexual risk-taking, in particular when the risk subjects considered are teenage girls.

Along with sociological perspectives on risk, feminist theories also informed the research design. Feminist researchers argue that feminist research is research 'for' women, rather than 'on' women. For some researchers sometimes what is supposed to be feminist research actually contributed to the perpetuation of the dominant androcentric research because of mistaken methodological choices¹¹. Referring to this type of inquiry as research 'on' women, Klein¹² argues that it is "often conducted without careful examination of the suitability of the methods used for feminist scholarships and the researchers do not state why they chose a particular method and what problems occurred during the research project". For the author, this results in research that may make women visible but not in "a feminist frame of reference"¹². She stresses that in order to use this frame it is necessary to employ research methods that take women's experiences into account.

Why is research on HIV/AIDS and other sexual risks a sensitive research?

Sensitive topics are not new on feminist researchers' agenda. Feminists have characterized 'sensitive research' as research in which the revealing of the 'private' to the 'public' is an expected outcome. Research on marital rape is considered by its author as 'sensitive' because of the dangers faced by women who share their private sexual experiences with the researcher¹³ Another study is taken as 'sensitive' because it investigates women's accounts of family lives, which involves the uncovering of private information¹⁴. In a report of a research project on gendered notions of pain the researcher highlights the sensitivity of pain perception, for it involves emotions and feelings¹⁵. In the case of my research, the 'private' to be revealed by the girls was also expressed with a mixture of emotions and the information disclosed could represent a threat to their moral integrity.

A research is sensitive when it can be threatening to its subjects¹⁶. 'Sensitivity' is a subjective construct, in the sense that "what feels sensitive or threatening to one may not to another"¹⁷. It should then be clear who is judging the research as 'sensitive', what is being considered as 'sensitive', and to whom the research is 'sensitive'. For the girls interviewed in the focus groups (FG), to express personal positions about sex meant to release private information about sexual knowledge and experiences into the public space of the meetings. And because the research involved pre-existing groups whose members had a continuing relationship before and after the fieldwork and within and beyond the context of the research, this could signify the danger of seeing secrets being spread out into the community.

Issues surrounding sexuality are commonly taken as 'sensitive' topics¹⁷. Power and gender relations, the question of sexual identity, the social and cultural norms that organize sexual behaviour, race, religion, age, etc. are all social determinants of the 'sensitivity' of talk on sexuality. This 'sensitivity' is ultimately constructed according to the norms and taboos of a given culture¹⁷.

Focus group research about views on sexual risks can also be considered as 'sensitive' because to talk about 'risk-taking' may involve the disclosure of perceived moral failures. The negativity of risk-taking is originated in the contemporary emphasis on the moral accountability for personal welfare. Health risk is understood as located within the self. It is seen as the result of a personal incapacity to take care of the self. This can be particularly problematic for research informants when the risk one talks about is recognizably a health risk (like HIV/AIDS) and the person who is supposed to guide the conversation is, like me, a health professional. Although 'risk' is not commonly taken as a sensitive research topic, some authors have already recognized that, in certain research settings, research participants can feel inhibited to disclose stories involving personal risk-taking because others can disapprove it.

The research design

With all this in mind I designed a qualitative and exploratory research to investigate teenage girls' ways of seeing HIV/AIDS and other sexual risks. The study was approved by the Research Ethics Committee at the Conceição Hospital Group, located in Porto Alegre, in Southern Brazil. The data were collected through FG interviews. The research procedure consisted in two sets of FG sessions with two groups of five girls. Each group had twelve weekly meetings, lasting approximately one hour and a half each. The girls signed a free informed consent form before participating in the research.

FG is an informal and semi-structured group interview that is carried out with the aim of collecting data on a specific topic of interest¹⁸. The reliance on the researcher's interests and the emphasis on group interaction are considered to be the two main features of the technique¹⁹. As FG researchers²⁰ point out, FG is "... ideal for exploring people's experiences, opinions, wishes and concerns". They also argue "the method is particularly useful for allowing participants to generate their own questions, frames and concepts and to pursue their own priorities on their own terms, in their own vocabulary"²⁰.

The choice of FG was due to the sensitivity of the topics to be discussed. The disclosure of information about sexual experiences or beliefs and attitudes towards sex and risk-related issues involves revealing intimate data. Although some might argue that a group environment is not ideal for inquiries involving very personal data, for it can be embarrassing to make it public, my research demonstrated that group meetings can work well when the research informants are teenagers. Given the sensitivity of the research, it was assumed that a degree of intimacy and friendship among the research participants and between them and the researcher would (and in fact did) facilitate the conversations. So, the decision was to use FG and, as discussed in the next session, to have research informants who knew each other previously. In the end, the group interaction resulted from that worked as a research facilitator.

Another reason to choose FG was the desire of making the data collection process enjoyable and somewhat useful for the research participants. The intention was to include as an explicit research interest the development of a research process 'for' the teenage women who participated in the research. Because FG allowed spontaneity, the research participants were able to discuss the topics, directing the conversations towards their own interests. As a consequence there was a reduction of the gap between the researcher's power and the power made available to the girls as research participants. It would be naÏve to affirm that this had the effect of eradicating the power differences inherent to researcher-researched relationships²¹. What was possible to achieve was a partial opening of the research process to the research participants' guidance.

The option for FG was also based on my desire to create within the research limits a forum for individual and collective reflection. I assumed that this was not commonly available to girls, either in research contexts or in the more natural settings of everyday life. Girls' narratives of personal sexual stories or expression of points of view about sex-related issues are traditionally under adults' surveillance. This has the effect of reducing the spontaneity of girls' conversations on sexuality, limiting the possibility of sharing of experiences. The less directive character of the FG technique offered the girls a chance to expose and share more freely their ideas, and time and space to rethink and change points of view before expressing them.

FG has already been recognized as a research method that can fulfill the feminists' goal of doing research in which research participants are not mere 'objects' of the research. As some researchers have argued, "focus groups are a relatively non-hierarchical method: that is, they shift the balance of power away from the researcher towards the research participants"²².

A third motif to choose FG as the main research procedure was its reflexive potential. In a FG "group members can describe the rich details of complex experiences and the reasoning behind their actions, beliefs, perceptions, and attitudes"¹⁸. In a group conversation both the researcher and the researched can, according to their interests, shift their positions from passive observers or listeners to active members of the group discussions. By being allowed to keep some distance from the research setting, both have the opportunity to reflect on individual experiences or assess personal beliefs about the issues that are being raised. This may have implications in the 'quality' of what is being asked and the answers it produces. If we consider sensitive research we can also argue that this feature of FG can be used in the interest of research participants because if the discussion is somehow embarrassing it can be up to them to talk or to remain silent.

In any research the dialogue between the researcher and the researched is quite important. Also important is the researcher's ability to listen to the research participants, in particular when the research deals with sex-related issues. In this sense the use of FG can facilitate the exploration of meanings and processes of meaning-making in HIV/AIDS-related inquiries. Group work may "ensure that priority is given to the respondents' hierarchy of importance, their language and concepts, their frameworks for understanding the world"¹¹.

The groups' constitution

A non-probability technique of sampling was employed to select the research participants on the grounds of research interests. For instance, because of the sensitivity of the issues to be discussed during the sessions it was in the interest of the research to have informants who shared a similar sociocultural context. To have something in common with other members of the group facilitated the participant's engagement in the discussions. It should be more comfortable to share ideas with recognized peers than with strangers. Another criteria for the selection of the informants was age. Giving that the main issue of the research was sexual risks, older girls were given preference. 14-17 year old teenage girls constituted the groups.

I did not select the research participants directly. I just established the geographical context of the fieldwork and advertised the research in two public schools situated in the area. My purpose was to work with working class teenage girls who were genuinely interested in participating in the research. The choice of working class girls had to do with my commitment to listen to girls whose views on sexual risk matters are more likely to remain invisible in the generalizing discourses of biomedicine. In the two schools from which I drew my groups I could not find more than six girls in each interested in participating in the research.

It has been argued that in FG research there is no rule for the number of group sessions as it will depend on the characteristics of the research, including time and resource limitations, and the researcher's interests¹⁹. Considering the limited amount of time I would have to carry out the data collection I elaborated a plan for 12 group meetings.

Given the characteristics of my research, I decided to limit the groups I would work with to two. I thought that two groups would be ideal due to the limited amount of time I had to collect the data and also the hard work involved in being a FG researcher. I do not consider the small number of groups to be a weakness of my research design. On the contrary, it was this feature that allowed me, I think, to keep a high level of involvement with the groups and to explore the research topic in the expected depth. As some researchers highlight, "statistical representativeness is not the aim of most focus group research"²⁰.

As it was commented earlier, a common feature of each of the two groups I worked with was that the group members knew each other very well. They studied in the same classroom, lived in the same neighborhood and/or also belonged to the same friendship network. The literature on FG research highlight that one of the important decisions to be taken by a FG researcher is whether or not to work with people who already know each other²⁰. Pre-existing groups have the advantage of being constituted by individuals who are, in one way or another, bound up with each other "through living, working or socializing together"²⁰. The importance of those types of social networks for FG research is that "these are, after all, the networks in which people might normally discuss (or evade) the sorts of issues likely to be raised in the research session and the 'naturally-occurring' group is one of the most important contexts in which ideas are formed and decisions made"²⁰.

To work with girls who knew each other well was good because it guaranteed the level of intimacy necessary for the discussions of the sensitive topics of the research. It also contributed to the level of homogeneity of the groups; 'homogeneity' in the sense of the sharing of a very similar sociocultural background. The literature considers 'homogeneity' in FG composition as a means to facilitate the flow of conversations within groups and also the analysis of the differences between groups²³.

Research setting and general format of the meetings

The chosen area for the development of the fieldwork was a working class area of Porto Alegre, Rio Grande do Sul, a state located in the south of Brazil. Even though I decided to recruit the research participants from two local public schools, I did not want to conduct the group meetings within the physical space of a school. My desire to avoid the school settings was based on the assumption that the context influences the data produced. I was thinking about the sensitivity of the topics to be discussed and the potential threat represented by the school environment to free discussions about sex. I wanted independence from school rules and bureaucracies. I also did not want to 'contaminate' the girls' conversations with the formality of the schools' environment. I ended up by conducting the meetings in a room that was used by the community for functions.

It has been argued that there are advantages and disadvantages in working with FG in informal settings²⁴. Some researchers²⁴ have commented that in their experiences, group conversations conducted in informal settings were 'chaotic' and more difficult to transcribe than those occurred in formal settings, because of the amount of interruption and unstructured ways of talking. They point out, however, that this is not necessarily a negative point as it may exemplify how real life interactions are constructed²⁴. I cannot say that the informality of the research setting resulted in 'chaotic' discussions. The girls themselves contributed a lot to the order of the group discussions. They helped me to get a balance between informality and order.

Each FG was constituted of three moments: an 'integrating activity'- IA, an interval with snacks and a focused discussion. I had separated weekly meetings with each of the two groups. The FG meetings were planned in advance, but the plans were relatively flexible. The literature recommends that FG research should be preceded by a phase of preparation in which the researcher studies the research topic and develops a protocol of broad concepts and guideline questions to be explored in the group discussions^18,20. Before starting the data collection I had a plan for the fieldwork, in terms of the general format of the meetings and the orienting themes for the first five. I did not prepare a plan for the rest of the fieldwork because of my wish to work with a 'stratified fieldwork plan', that is to say, a plan that was going to be constructed step by step as the data collection progressed. I organized an initial agenda with questions and topics that I considered relevant to be answered and/or discussed. However, I was prepared to re-think the agendas during the course of the data collection, as other relevant topics should emerge from the group interactions (for the final focus groups plan see Box 1).

The way I developed the fieldwork plan demanded a high level of personal involvement with the research. After each meeting I had the task of listening to the focus group recorded tape, writing down in my diary some notes about what would possibly be important to discuss in the next focus groups and why. This procedure allowed me to develop a reflexive interaction with the data from the earlier stages of the data collection process onwards. The next task was to compare the topics listed as relevant with the previous agenda and decided what was going to be the topic explored in the next focus group meeting. As I worked with two groups separately, I had to follow those procedures at least twice a week. This proved to be a demanding and time-consuming task, but also a rewarding one. The more I opened to the participants the direction of the focus group discussions, the more I was able to approach the 'reading grid' they use to make sense of the risks of sex in general, and HIV/AIDS risk in particular.

The organization and analysis of the FG data

The group meetings were tape-recorded, a procedure authorized by the informants. All the recorded focused discussions were transcribed and used as the material source for data analysis. Data were analyzed through an ongoing process and occurred concurrently with and immediately after data collection. The transcriptions were analyzed in two stages - data coding and data interpretation. All mentions of a given theme were coded either originating from individual or group-produced discourses. The coded quotes were grouped into categories and sub-categories, a procedure oriented by the research questions. That operation resulted in the interpretative framework upon which the analysis was developed.

Integrating activities (IAs)

FGs are social events (although not 'natural' ones) that are usually enjoyed by its participants¹⁸, especially if the group members are interested in the topics to be discussed. That is the case when the research informants are teenagers and the research topic is sexuality. Although the interactive nature of FGs is expected to make the informants' participation in FG research more enjoyable than in other types of studies, it is still possible to enhance their pleasure and involvement with the research. To start each meeting with an integrated activity (IA), a sort of 'warming up' moment where the group members informally interact can contribute a great deal to the group interaction and making the meetings enjoyable.

What I called IA was a sort of 'warm-up' group activity, lasting approximately 30 minutes and designed with the primary purpose of making the group meetings as enjoyable and interactive as possible. Secondly, the IA was also thought of as a means to facilitate the participants' involvement with the research and to generate focus for the group discussions. In the end, what was experienced in an IA - the stories created, the jokes, the language used, the roles played, the insights, the issues raised - became substantive material for the FG discussions.

The idea of using IAs to begin the group meetings was based on the assumption that "...the point of doing group interview is to bring a number of different perspectives into contact"²³. In this sense, the use of an IA to provide space for the participants to interact with others on a given topic had the advantage of stimulating the awareness and defense of personal positions before the actual focused discussion took place²³. The differences between personal positions started to emerge and became clear to the participants themselves during their involvement with the IA.

Conducting the group discussions

In FG research on sensitive topics the style of the group moderator is crucial²⁵. On the one hand, it is obviously important to avoid asking highly personal questions; on the other, to marginalize some issues simply because one assumes that no one could possibly want to disclose private experiences to other group members is also ethically problematic. One way of dealing with this dilemma is, the authors point out, to let sensitive discussions to be maintained by the research participants rather than the researcher. So, it was not the case that sensitive questions were off my agenda but that I tried to formulate them in a way that would not directly implicate the 'self' with them. If the 'self' became visible it was because the participants themselves were interested in making 'the personal' visible. Despite my intentions, I cannot be sure that this was always the case. If the researcher is very much involved in the group interaction, like I was, it is difficult to always maintain a certain level of rationality and self-consciousness about how the decisions should be taken or which rules they were supposed to be following. In my particular case, the flow of the FG discussions was very often their own regulator.

I took advantage of the interaction created in the IAs and in the subsequent 'interval' (where we used to have snacks, sometimes brought by one of the participants, sometimes by myself) to extend the informality and spontaneity to the third part of the meeting. All the sessions were conducted in a relaxed way. I tried to interfere as little as possible in the groups' discussions. My role as a facilitator included reminding the participants at the beginning of the session about the theme to be discussed. I sometimes used a question or a statement related to the original theme but produced in the IA to start the session and then left the participants free to embark on a conversation about it or to present their points of view.

As a facilitator I could not be passive, for it was not always possible to leave the discussions to the exclusive guidance of the participants. A more "interventionist style"¹⁹ was used to stimulate the continuing exploration of a given theme, knowing that this would provide relevant information for the research. At other times I simply brought the discussion back to a theme that had been emphatically discussed before but in a superficial way. When intervening in the group's interaction, I was attentive to the participants' reactions to my positions or to my interpretations of their positions. I usually called for a more explicit manifestation of those reactions. This had the effect of maximizing the interactions between the participants, for to explain a reaction included pointing out its reasons and presenting to others an exploration of personal arguments. Once the interaction started to be productive, I used to return to a less interventionist style.

With regard to my involvement with the discussions, I adopted a position where I was not totally open to disclose personal experiences and ideas to the research participants. I was continuously worried that my participation could inhibit their manifestations, in the sense of setting unintended limits to it. So, there was always a tension between my conscious presence within the group and how much of me could be revealed without compromising the presence of the participants and the consequent production of data.

Each FG session had an orienting theme. Before the session I always had a short list of the topics I would like to explore. However, this initial agenda was flexible enough to be totally or partially changed in order to accommodate new questions or topics of interest brought from the group interactions during the IA of the day.

Concluding comments

In this paper I have argued that FG is a good methodological option for researchers who intend to carry out studies on sensitive topics, such as HIV/AIDS, sex and risk. At first sight the technique can be taken as inadequate for sensitive research. My experience demonstrated, however, that, as with any other research tool, the potentials of FG for qualitative health research and, in particular for HIV/AIDS research, depend on research interests and on how the fieldwork is planned and conducted. In the specific case of my research, I considered that the aim of investigating meanings and processes of meaning-making through teenage girls' own accounts of the risks of sex could not be fulfilled with the same success if the process of data collection had not been carried out in a group environment.

In a group, providing the group members follow the principles of confidentiality, research participants may feel safer when it comes to revealing intimate information. The sharing character of the information disclosed in-group discussions acts as a protecting factor, in the sense that what is revealing results from a sort of collective construction. In the specific case of sensitive research involving teenage informants, I would say that the great advantage of FG is to provide a research context that is less dominated by the adult who happens to be the researcher, giving the informants the possibility of choosing when and how to participate in the research.

Sensitive research demands, perhaps, more sensibility on the part of the researcher. There is the need of a constant awareness of the links and impact of the process of data gathering upon the researcher's and research participants' feelings and lives.

I did not intend to discuss here the issue of the use of FGs in sensitive research in its full. The theme is complex and I do hope to have motivated other researchers to reflect upon and write about the question of how to carry out research on sensitive topics in group meetings, particularly when the research participants are teenagers and the topics studied have to do with HIV/AIDS and other sexual risks.

Acknowledgements

I would like to thank the girls who participated in my research for sharing with me their experiences in such a sensitive terrain like that of heterosexual relationships. I would also like to thank CAPES - Federal Agency for Post-Graduate Education/Brazil, for its financial support.

Artigo apresentado em 16/07/2009

Aprovado em 09/09/09

Versão final apresentada em 01/10/2009

• 1. Siegel K, Schrimshaw EW. Perceiving benefits in adversity: stress-related growth in women living with HIV/AIDS. Soc Sci Med 2000; 51(10):1543-1554.
• 2. Hebling EM, Guimarães IRF. Women and AIDS: gender relations and condom use with steady partners. Cad Saude Publica 2004; 20(5):1211-1218.
• 3. Oliveira DLLC, Meyer DE, Santos LHS, Wilhelms DM. A negociação do sexo seguro na tv: discursos de gênero nas falas de agentes comunitárias de saúde do programa Saúde da Família de Porto Alegre, Rio Grande do Sul, Brasil. Cad Saude Publica 2004; 20(5):1309-1318.
• 4. Doyal L, Anderson J. 'My fear is to fall in love again...' How HIV-positive African women survive in London. Soc Sci Med 2005; 60(8):1729-1738.
• 5. Lesch E, Kruger LM. Mothers, daughters and sexual agency in one low-income South African community. Soc Sci Med 2005; 61(5):1072-1082.
• 6. Oliveira DLLC. Brazilian adolescent women talk about HIV/AIDS risk: reconceptualizing risky sex - what implications for health promotion? [thesis]. London: University of London; 2001.
• 7. Giddens A. Modernity and self-identity Cambridge: Polity Press; 1996.
• 8. Beck U. Risk society: towards a new modernity London: Sage; 1992.
• 9. Douglas M, Wildavsky A. Risk and culture Oxford: Basil Blackwell; 1982.
• 10. Heyman B. Introduction. In: Heyman B, editor. Risk, health and health care: a qualitative approach London: Arnold; 1998. p. 5-10.
• 11. Kitzinger J. Focus groups: method or madness? In: Boulton M, editor. Challenge and innovation: methodological advances in social research on HIV/AIDS London: Taylor & Francis; 1994. p. 159-175.
• 12. Klein RD. How to do what we want to do: thoughts about feminist methodology. In: Bowles G, Klein RD, editors. Theories of women's studies London: Routledge & Kegan Paul; 1983. p. 80-104.
• 13. Bergen RK. Interviewing survivors of marital rape: doing feminist research on sensitive topics. In: Renzetti CM, Lee RM, editors. Researching sensitive topics London: Sage; 1993. p. 197-211.
• 14. Edwards R. An education in interviewing: placing the researcher and the research. In: Renzetti CM, Lee RM, editors. Researching sensitive topics London: Sage; 1993. p. 181-196.
• 15. Bendelow G. Using visual imagery to explore gendered notions of pain. In: Renzetti CM, Lee RM, editors. Researching sensitive topics London: Sage; 1993. p. 212-228.
• 16. Dickson-Swift V, James EL, Liamputtong P. Undertaking sensitive research in the health and social sciences: managing boundaries, emotions and risks New York: Cambridge University Press; 2009.
• 17. Farquhar C, Das R. Are focus groups suitable for 'sensitive' topics? In: Barbour RS, Kitzinger J, editors. Developing focus group research London: Sage; 1999. p. 47-63.
• 18. Morse JM, editor. Critical issues in qualitative research methods London: Sage Publications; 1994.
• 19. Kitzinger J. The methodology of focus group: the importance of interaction between research participants. Sociol Health Illn 1994; 16(1):103-120.
• 20. Kitzinger J, Barbour RS. Introduction: the challenge and promise of focus groups. In: Barbour RS, Kitzinger J, editors. Developing focus group research: politics, theory and practice London: Sage; 1999. p. 1-20.
• 21. Maynard M, Purvis J. Researching women's lives from a feminist perspective London: Taylor & Francis; 1994.
• 22. Wilkinson S. How useful are focus groups in feminist research? In: Barbour RS, Kitzinger J, editors. Developing focus group research: politics, theory and practice London: Sage; 1999. p. 64-78.
• 23. Morgan DL. Focus groups as qualitative research London: Sage; 1997.
• 24. Green J, Hart L. The impact of context on data In: Barbour RS, Kitzinger J, editors. Developing focus group research: politics, theory and practice. London: Sage; 1999. p. 21-35.
• 25. Kitzinger J, Farquhar C. The analytical potential of 'sensitive moments' in focus group discussions. In: Barbour RS, Kitizinger J, editors. Developing focus group research: politics, theory and practice London: Sage; 1999. p. 156-172.

The use of focus groups to investigate sensitive topics: an example taken from research on adolescent girls' perceptions about sexual risks

O uso de grupos focais para investigar temas de natureza íntima: exemplo de uma pesquisa sobre pontos de vista de meninas adolescentes acerca dos riscos do sexo

Publication Dates

• Publication in this collection
  21 July 2011
• Date of issue
  July 2011

History

• Received
  16 July 2009
• Accepted
  01 Oct 2009