ARTICLES

 

Examination and analysis of researcher-participant power dynamics in focus group discussions

 

 

Mzikazi Nduna

Department of Psychology, University of the Witwatersrand Private Bag X3, Wits University, 2050 Mzikazi.Nduna@wits.ac.za

 

 

---

ABSTRACT

Focus group discussions (FGDs) are a popular method of collecting data for various qualitative research study designs. The value of focus group discussions (FGDs) in community-based studies is enhanced with rigorous application of the research method. Its flexibility enhances its usefulness in varied contexts, research questions and disciplines. However, power dynamics could affect optimum use of the method, the kind and amount of data collected. Based on research experience mainly in South African communities; villages and townships, this paper explores the performance of power in FGDs. This paper discusses researcher-participant power dynamics in community engagement, participant recruitment, organisation and facilitation of FGDs. The experiences discussed here are applicable in a variety of research topics, and beyond the South African context.

Keywords: ethics, focus group, power, South Africa

---

 

 

Introduction

This paper presents a critical reflection from the author, of some of the successes and pitfalls encountered in recruiting for, and conducting face-to-face focus group discussions in rural and urban settings in South Africa. The insights provided here are useful for the transformation of teaching and research undertaking that is called by scholars in this special issue (Ramose & Baloyi, 2020). This reflection contributes to the body of work from South Africa that brings the voices from the South to the growth, development and of qualitative research methods (Macun & Posel, 1998). FGDs can be used in a study as a standalone method or as part of method triangulation together with interviews, direct observation, participant observation, document analysis, archival records and examination of physical artefacts (Macun & Posel, 1998; Mnguni, 2016). FGD data can also be used to corroborate previously gathered data. This paper is based on the author's experience with using single and mixed-sex groups discussions, across age groups, in the various provinces in South Africa, and conducting generic training of fieldwork teams for various organisations. This article reflects on, and make suggestions on how to avoid some of the challenges and to better plan for qualitative studies under similar conditions. The paper is divided into three sections: the discussion of research ethics, community mobilization and engagement with science, ways to organise and facilitate a FGD.

Focus Groups are one of the various tools used to collect data in qualitative studies (Macun & Posel, 1998). The style of how the group is organised can be unstructured, semi-structured or structured. A structured group would follow a formalised process which could be duplicable. An unstructured group would observe group discussion rules but be less formalised in its formation and conduct. In a semi-structured group discussion some elements of structure will be there but with leeway for the facilitator to respond to the organic process of the group lifecycle. Recording of data is done through note taking, audio or video recording depending on the type of data needed to answer the research question.

There are two main types of focus group data collection methods: the interviews and discussions. This paper is concerned with the facilitated, face-to-face, focus group discussions (FGDs). Different conceptualisations of focus groups range from face-to-face to internet-based focus groups which are facilitated by a researcher or self-directed (Morison, Mtshengu, Sandfort, & Reddy, 2016; Moroke & Graham, 2020). Competencies and qualities of staff needed for the study depend on the nature of the study. Within a single project, team composition may require multiple skills, matching of a facilitator to the group; and facilitators with a firm grasp of literature in the research area being studied and be unbiased by preconceived notions (Nduna, Sikweyiya, Khunou, Pambo, & Mdletshe, 2016).

Focus group discussions offer benefits to researchers because they save time and allow the researcher to access a greater number of participants in a shorter space of time than one-on-one interviews (Moroke & Graham, 2020). In line with qualitative research, the groups are kept small: between six and ten participants, whereas some researchers utilize up to 40 participants (Macun & Posel, 1998; Oktay, 2004). It is important to plan for disappointment when recruiting participants and recruit slightly more than the number needed to ensure enough participants (Mavungu, Thomson-de Boor, & Mphaka, 2013). Some volunteers may not fit the recruitment criteria well but are keen to partake in the study. The researcher, guided by their inclusion and exclusion criteria for the study should screen all volunteers for eligibility and identify participants who experienced the phenomenon under study.

The description in this introduction does not suggest that this method is applied seamlessly; hence this paper offers a reflection of some dynamics with implementation. When recruiting participants who fit specified narrow criteria for the study it is easier to use referrals, existing civil society organizations and support groups. To reach broader community members, depending on the target group, newspaper advertisements, community radio presentations, community meetings (in schools, churches, clinics, chief's place etcetera) are used and are useful where the researcher is not looking for representations. The usefulness of each of these recruitment strategies depends on the setting and researchers need to consider the merits and demerits of each. FGD elicits group dynamics which yield more data and questions that the researcher would not have thought of. FGD enhances trustworthiness and may minimise the opportunity for participants to exaggerate and misrepresent their responses as peers in the group could identify any such attempts and challenge them.

Depending on the type of data that the researchers seek to obtain, different interviewing approaches are used. The interview guide can comprise open-ended or focused questions. In standardized FGDs questions are decided upon in advance and asked with the same order. The Interviewer does not have flexibility to reword questions, introduce new questions or change the order of topics. Standardized questioning minimizes errors of questions, yields reliable and comparable case-to-case data, best suited for focus group interviews. There is a list of objectives, which informs the interview guide with a set of suggested questions, because without questions, the facilitator might overlook asking some important dimensions. The interviewer has latitude and flexibility within the framework of the interview guide. The wording can vary from one group to the next, questions can be reworded, and the facilitator can introduce 'new' probing questions and change the order of topics but keep in mind the objectives. In unstandardized FGDs the facilitator is completely flexible. To conform to the respondents' spontaneous sequence of ideas and idiosyncratic experiences, the discussion can vary from one group to the next. An unstandardized interview approach encourages true-to-life responses based on these idiosyncratic experiences. This bears in mind that, even though the participant may face the same situation they may not perceive or manage it in the same way as another participant in the group or as the facilitator would. However, data saturation is reached when no new experiences are shared.

 

Discussion

In qualitative research, ethical dilemmas and decision-making in relation to all aspects of the study are negotiated from moment-to-moment throughout the study.

Application of ethical principles is peculiar when conducting focus groups discussions. Concerning confidentiality, the facilitator should encourage confidentiality but be clear that this cannot be guaranteed. Perhaps what helps would be to introduce the Chatham House rules¹ and explain the risks of disclosure to the group. Chatham rules encourage that participants are free to use the information received, but neither the identity nor the affiliation of the speaker(s), nor that of any other participant may be revealed. Throughout the data collection, the researcher needs to be mindful of the conflicting interest between the imperative to balance participant's safety and the pursuit of science as this is an important ethic of care.

The facilitator needs to be familiar with the values of the study community and how respect is afforded to different people and processes. The microdynamics in the field and moments of productions of power require respect of differences. The next three scenarios reflect moments of field worker power to decide on fairness and justice, beneficence and respect.

Firstly, respect for the community could be afforded by demonstrating tolerance to shared community values even when these are not values that the researcher ascribes to. The tension between secularism and religion presents wherein the group wishes to start with a prayer; the facilitator cannot enforce this if it is their value but not that of the group. The researcher may not necessarily agree with some of the values and practices of their target group, but they may need to concede.

Secondly, experimental ethical dilemmas pose unexpected situations during fieldwork and the facilitator has to respond to these as and when they present. For instance, in one FGD of young adults it was reported by the field worker that one of the participants coughed insistently with what the field worker suspected was tuberculosis. TB is an airborne disease. The facilitator worked at her local clinic as a peer educator for TB High Transmission hot spots and was acutely aware of the signs and symptoms of TB having survived TB herself. In this case, the field worker could not exclude the participant, as that would have been discriminatory. She experienced an ethical dilemma. Potential exposure to a communicable disease could pose a risk of infection for other members, when it was unknown to the facilitator if the suspected participant was on treatment. When a typical risk dilemma presents, there is no research supervisor, no principal investigator, and no Institutional Review Board Committee on site to give guidance.

A third example of an ethical dilemma and a threat to the trustworthiness and credibility of the data relates to mental illness. Mindful of the fact that the proportion of people with undiagnosed mental health conditions may be high in rural settings, mental ill-health as an exclusion criteria at recruitment may not always work. When the facilitator discovers during the discussion that a participant should have been excluded, should the participant be asked to excuse themselves? In all three scenarios, the field worker has the power to decide, include and or exclude.

Community mobilisation ensures that the community is educated about the scientific exercise, the purpose of research, sampling and other aspects of the research process and give feedback about research findings. Prolonged community engagement keeps the research community abreast, consulted, and knowledgeable about research. This constitutes an ethically responsible way of conducting research. Without this education, some volunteers might come to the FGD thinking that it is a workshop where they will learn.

For instance, here is what one participant in a study conducted in KZN seemed to think about the research:

Mr S.F: "...Maybe the facilitator and the researchers want to hear our opinion on these issues...Our viewpoint is needed because it must go to the government so that the government should have plans for these issues we are speaking about.."

Mr S.B: "...but. we are few to discuss such issues it would be better if we fully packed this hall as we are being taught."

Mr S.F: "We are not being taught Baba we are voicing out our viewpoint.."

Research projects should budget for prolonged and ongoing community engagement and sometimes beyond the data collection and publication of study findings. The study information brochure is not always sufficient for this purpose.

Mr S.B: "What about those who are left, behind? When will they get a chance to voice out their viewpoint?"

Where participants do not understand sampling such a concern may ensue and derail the process hence efforts at creating awareness and knowledge about science are key.

Once entry to community is granted, the researchers recruit volunteers. In a community generally known for dealing with illegal dagga trade (Ewing et al, 2013) young men refused to participate in one of the studies that I was involved with, citing suspicion that the research was linked to the government. The field workers for this site tried to secure an appointment at least four times and respected the collectives' refusal to participate. This has budget implications for the researchers' repeated visits to the site. Sometimes prospective participants refuse, exclude themselves midway and or fail to show up. No matter how limited, participants' decision to partake in studies and how they wish to be part of the study is in their power and they should always be allowed to exercise this (Kessi, Kaminer, Boonzaier, & Learmonth, 2019; Macun & Posel, 1998); this is despite the glaring power scale tilted in favour of the researcher in most cases. When appointments are not honoured, participants' behaviour could be taken as an exercise of their right to withdraw; there is no need to interrogate participants about it. There were other instances where political interference from the local ward councillor made it impossible to proceed with recruitment and where field workers were chased away. Some of these challenges, which are related to shifting nexus of power between the researchers and the community are possible to avoid if unassuming and unrushed community mobilisation is thoroughly undertaken and involves all the key community stakeholders. At the same time community stakeholders may gate-keep the process; miscommunicate the research and bias recruitment to leverage other gains.

When competing interests on the participants' side occur, there would be late arrivals and people who leave the group before the finish time. Double commitments for participants negatively affect maximum and meaningful participation. Because some of the reasons for changes in plans, such as death, are impossible to plan for. Some villages have mourning rituals and pragmatic burial clubs and societies that involve a collective response and attendance to bereavement. If a FGD appointment coincides with such activity; it may be cancelled. Splitting attendance of the community between two equally important meetings is undesirable and not cost effective. Otherwise continuing with the process will exclude some eligible participants and negatively affect the sample, bias the data and waste research resources. Factors such as age, gender, employment status, commitments such as childcare, church attendance, coop projects et cetera affect facilitation and participation in groups. Women caring for children cannot attend or bring their small children. In the three-hour FGD some mothers may need to attend to feeding, changing nappy or other needs of the child. Hence, a break could be necessary. Sometimes the child(ren) may cry and seek the mother's attention and this distracts the process.

Venue availability is also depended on other community activities which may take precedence. A researcher familiar with the local culture would prepare for these, have flexible time frames, budget to cover costs associated with the inconvenience and negotiate decisions around these. Projects with very tight budgets and time frames do not allow for contextually relevant and culturally responsible data collection and create conflict for researchers.

 

Power dynamics in FGD facilitation

The need to appreciate and interrogate power dynamics in qualitative community research is discussed by other scholars (Kessi et al., 2019). Reflexive scholarship requires and enables the researcher to consciously and consistently be aware of the power shifts in research. This is in relations to the researcher and the participants, participants themselves and participants and other stakeholders whose world is accessible to the researcher through the participants (Kessi et al., 2019). All of these are examined here. The behaviour of each member of the group is determined by perceptions of the other in terms of status, power position, and probable views. Certain kinds of communication occur most easily if the respondents perceived themselves to be higher, lower or same level as they perceived the facilitator on the social scale; hence matching facilitator to the group is ideal (Oktay, 2004). People are generally happier and more anxious to speak to those above them in the hierarchy than those below them. On the other hand, the person of a lower class may be motivated to present themselves in a favorable light to the researcher if they perceived them to be influential or they may express some elements of hostility based on resentment of their social conditions. Facilitators need to be aware of the - isms: sexism, racism, xenophobia, heterosexism, fundamentalism and their contribution to power. These can create serious problems for participation.

 

Gender-power dynamics and safe space in focus group discussions

Conducting paired or matched (to the facilitator) peer groups is ideal to achieve maximum participation, control dominance, and create a safe space for both the group and the interviewer. Considerations for matching, in each study, depend on the context, the sample and the research question. It may be prudent to consider demographic background and participants' experiences in relation to the topic under study because these are often areas of intersection and positions on which sometimes participants may vehemently differ with each other. For example, men and women may have ideas and experiences that they wish to share in a safe space and not risk being judged by the opposite sex. These are experiences such as transactional sex, rape, sexual dysfunction, and intimate partner/domestic violence. Sometimes differences of opinions, experience and positionality can introduce subtle or obvious undesirable performance of power(lessness) in the group. Below is an example of how gender can affect facilitation and participation. In an HIV/AIDS study where the facilitator was a young man and the group was mixed sex, he noticed that "... On the female side they are quiet..." (Silence)

These are the kind of imposed and awkward silences that are observed in a cultural context where women do not engage with men on debate. Here is another example to demonstrate this point. A woman facilitated this mixed-sex focus group discussion and gender power was demonstrated in this way:

Facilitator: "Let's first hear what female 6 wanted to say then you can put your argument forward!"

Female participant 6: "He interrupted me, my confidence is low!" (Few people started talking together at the same time and laughing)

Facilitator: "You can talk participant 6; we heard you wanted to say something!"

Female participant 6: "He interrupted me and shattered my confidence that is what I would say, let me put it like this I see that he has something against women... I see there is still a lot of suppression!"

Facilitator: "When you say women are suppressed, please elaborate because you said he is against women, what kind of suppression is it elaborate!"

Female participant 6: "He is emotionally blackmailing me (laughs)!"

It would seem that the participant felt disrespected and dismissed, and this resulted in disinterest to continue with the discussion on her part and thus withholding potential data. This vigilance on the facilitator's side is essential especially if monolithic peer group composition was not possible. Some male participants may express misogynistic attitudes, infantise, attack, belittle and intimidate female participants. When participants feel that, their safety, even emotional safety, is compromised and this may result in silence and a sense of worthlessness. This may result in loss of data for the researcher. Conflicts around lack of co-operation, disregard for work ethic and unprofessional behaviour between facilitators may affect data collection, lead to delays, waste resources and result in poor quality of the data. Conflictual co-worker dynamics negatively interfere with work and staff morale - for both the reporter (the corporative) and the reported (uncooperative). Pairing of co-workers would ideally take into account similar sources of power as in matching for group facilitation: age and gender.

 

Age and power dynamics

Matching on the basis of age, peer groups may be tricky in communities where things other than age, such as having a child, being married, education level, initiation status, et cetera matter (Spiegel and Mehlwana, 1997). In another study an older male facilitator was uncomfortable talking about HIV and he reported in his field notes that it was because he knew '...there was one guy infected by the disease..I The facilitator's discomfort prevailed despite having undergone a one-week of training on HIV/AIDS research. The facilitator was from the same location as the participants. Further in field notes about the process, he noted that '...luckily the guy went out and then I started the section.. I which he had skipped deliberately to protect his discomfort. It was unclear why the participant left the room. When the facilitator was asked by the research team about his discomfort, he suggested that it was more about the participant being older and he thus needed to show him some respect, than about the discussion of the subject matter in his presence.

Elsewhere, young men and women who were postgraduate students conducted FGDs about sexual harassment with men and women general workers on campus (Kiguwa et al., 2015). Participants and facilitators were all Black and African and at debriefing meetings facilitators reported that they experienced discomfort when facilitating the discussions because some of the participants were old enough to be their parents.

 

Shifting power dynamics during facilitation

Power dynamics may shift in research interviews. The article by (Moroke & Graham, 2020) demonstrates this in recruitment and interviewing. The facilitator should take charge and not allow one person to dominate the discussion. A facilitator should not allow participants to take over the facilitation. For instance, participants should not tell the facilitator to move to the next question - that should be the judgment call of the facilitator. The facilitator should allow for a more robust debate, not take only one or two answers, and move on. Silences are permitted and sometimes necessary as they give an opportunity for participants to reflect, organize their thoughts and contribute. Research participation could be seen by participants as a powerful opportunity to represent oneself (Kessi et al., 2019), therefore participants will speak from personal experiences or general knowledge; the facilitator should allow for this and provide the necessary support in response to personal disclosures. Participants represent the diversity that exists in the communities and hence in a group they will agree on some things but not on others. The facilitation should allow for behavior that exhibits that which is found in natural contexts such as moments of sadness, anger, excitement, disagreements et cetera but make sure there is no attack on the person.

Managing participants power dynamic is about allowing participants to make decisions about how they would like to participate in a study (Kessi et al., 2019). In situations where participants will know certain things about each other, the facilitator should discourage participants from making examples of each other in the group, speaking on another's behalf should also be discouraged. For instance, if a participant withholds 'known' information, that should be respected by the group. This person's story should not be told on their behalf without their consent. Each participant's autonomy in the group should be respected. The facilitator should take charge and not allow participants to dismiss each other. An example of a participant who has taken over facilitation is one who tells another participant to 'shut up' or tell them 'you do not know what you are talking about', et cetera.

Side talks will happen, in pairs or small groupings. If participants talk at the same time the facilitator should wait for the group excitement to subside and ask each person, one-by-one to repeat themselves so that their input is captured on the audio-recorder and so that the facilitator can note down areas for follow up and need probing. Side conversations shall not be ignored as they may be in disagreement with the loud voices in the group and need to be given a special opportunity to be heard. Sometimes these happen because someone is shy to speak and express what they feel may not be a popular opinion. When a pair (or small group) talks amongst themselves they should be invited to share their conversation with the big group.

 

Correcting misinformation

Whilst the role of the facilitator is to gather research data, they could find ways to correct misinformation through the groups by probing for counter knowledge, by allowing participants to correct each other and ask each other questions without stepping out of their role as a researcher. In a sexual and reproductive health and rights (SRHR) study conducted in the Eastern Cape and Mpumalanga provinces participants were asked about early sexual debut. In two different sites which are 1,009.6 kms apart, respondents agreed that sexual debut started at:

Participant 6: "...8 years because of the food that is not well for their bodies, they no longer eat (dipudi); they eat bunnychaws which makes them sexual!" (Few people laugh) (FGD, older women, Eastern Cape)

Participant: "12 years (They all agree with participant Yes)...think (.), I think the food that they eat is very nutritious. It is the diet.." (FGD, older men, Mpumalanga)

This view about the relationship between modern fast foods and early sexual debut can be considered an urban legend that fits with established folklore theories. It is such myths that could be corrected. Another example is as follows

Participant 4: "I think I agree with female 8 when she say it (HIV) weakness the immune system quickly in women because they go through periods²..they have abortions... miscarriages..there can be so much happening in a woman's body ... that is why it weakens them quickly and they also protect the child that's why they are dying..."

Participant 5: "And the mother's blood goes to the child!" (FGD, Older men, Mpumalanga)

In an attempt to correct misinformation or perceptions of LGBTI persons the facilitator read the body language of another participant who expressed disagreement and called on him to make a contribution. The discussion ensued as follows:

Interviewer: "Here is Mr S.K maybe he will give an answer."

Mr S.K: "What I want to explain is that Bisexual is a person who loves both. He can be in love with a man and be in love with a woman... It can be a man who sleep with women and sleep with other men as well. And it can be a woman who sleep with women and sleep with men as well. But the one with both sex organs is called intersex."

Mr S.F: "What do you call that in Zulu?"

Mr S.M: "Uncukubili because he has a double." (AFSA Study, FGD, Eshowe, 2014)

Below are other examples from a sexual and reproductive health and rights (SRHR) study where the facilitator asked questions about HIV, unplanned pregnancies and abortion and misinformation was provided and corrected.

Mr S.J: "...you will be shocked to know that those who are not after money they abort and throw them [fetus] into the forest or in the cane field ...they strangle the baby and throw into the bush... we have many such cases here in our place and you can see that this thing is increasing!... Now people from the community will meet a dog running with an arm or a leg of an infant."

Mr S.A: "What I want to say is that, you see as we are here we from this community so I cannot worry myself with something which never happened, firstly it confuses my mind when you talking about meeting a baby, we never met any aborted baby here so it confuses me to be serious about something that never happened. What is important here is for you to tell us what must be done. We all neighbours here we never met any dog running with a babies leg... " (AFSA Study, Eshowe, 2014)

In addition to correcting misinformation, the facilitator could find ways to correct discriminatory and uncouth attitudes through the group's discussion. For example, the facilitator should not let attitudes like these pass:

Facilitator (male): "We heard you talking about (tuba dimpa) abortion, what is that?"

Participant 4: "Yes they have abortions; yes like now most girls don't, work so they need money... she will have an abortion you see. From there that person has dirt and it's like she has a mortuary in her stomach..." (Few females laugh)

This is the uncomfortable laughter that Macleod and Morison (2015) refer to in their study of pathways to parenthood. A similar discourse about multiple concurrent sexual partnering and transactional sex was expressed in a mixed young men and women FGD where some said,

Participant 4: "...so you see...these children /"referring to young women] need to understand that their bodies are diamonds; stop giving them away like they are water in the bushes!"

Facilitator: "When you say they are like water in the bushes, what do you mean?"

Participant 4: "I mean like water in the bushes welcomes anything you see, donkey drinks, a cow drinks, rabbit drinks, springbok drinks, lion drinks; you see!"

 

Asking the questions

It helps to ask the respondents when they speak for the first time to identify themselves and describe their role. For example, you may open by reiterating an identifying question. Such as 'tell me what you do, in what position and for how long'. This information helps the researchers with analysis and data interpretation later. Open questions generally state an issue without providing a list of answers and respondents will vary greatly in the frame of reference from which they answer. This allows the respondents to follow their own train of thought. The choice to ask questions in terms of the respondent's own immediate experience or in terms of generalities will depend on the study design and aim to achieve specificity, specify the time, place and context you want in your answer. Rather than 'what do you normally do...' you could ask 'what did you do the last time...' Rather than 'how much time do you spend on TV', you could ask '...how many hours do you spend on TV...' Keep the respondent from feeling that certain answers will mean loss of prestige. You may assume that the respondent possesses the low-valued attitude or behavior in some degree and place the burden of denial on them. For example, 'when did you first...', instead of 'did you ever...'?

The facilitator may ask unusual questions following up on what was said by the respondent. Obtaining and maintaining the respondent's interest for the duration of the discussion is essential. This can be done by paying attention to the way that questions are asked, probing, handling embarrassing questions and demeanor. Respondent's behaviour during the interview might signal their continued interest or lack thereof in the subject.

The facilitator should establish rapport but be careful of 'over-rapport'. Avoid forming close personal friendships with respondents otherwise certain questions become difficult to ask and some participants may exhibit dependency and seek advice, help, praise, et cetera. A tactful facilitator should not appear to reject these.

Use local language as much as possible (Mnguni, 2016): this is respectful and it will yield meaningful data. Make no assumptions about language competency based on professional status of the interviewee; people are not always comfortable and able to speak English. See example below where a professional KI preferred to speak their indigenous language.

Field worker: "(aha) okay, the youth (.) Why they are so vulnerable would you say, what makes them vulnerable to HIV needs, violence?"

Health educator: "By doing what should say (eish), (Laughs) It is difficult to use the English language to answer" (Sept/2012: DR JS Moroka district, KII)

The fieldworker asked a question and there was silence. He checked:

Field worker: "Do we understand the question?"

Participant 6: "May you please explain it in Setswana; I think English confuses me a bit". (Few females laugh)

Field worker: "Must I ask it in Setswana?"

Participant 6: "Yes!"

Field worker: "Okay you do not understand they ask what help and needs that are there for people who...Setswana is difficult let me put it in my understanding..."

The research from South Africa suggests that decisions around which language to use is sometimes influenced by the gate keepers or participant's preference to be interviewed in English (Moroke & Graham, 2020; Ngunyulu, Mulaudzi, Peu, Khumisi, & Sethole, 2016). Qualitative researchers take on the authority and responsibility to report the views of people with very different lives and views of the worlds. The effort to learn the internal views of others is worthy and important, and as a key task of qualitative researchers, it is always done with empathy. Qualitative researchers should 'walk in the shoes' of others and as the former president of South Africa said 'If you talk to a man in a language he understands, that goes to his head. If you talk to him in his language, that goes to his heart' - Nelson Mandela³, Former President of South Africa.

Contextualize embarrassing and sensitive questions; consider things like religious values, sexual orientation, history of abuse, and other experiences. It is desirable to preface questions with an explanation.and indicate that sometimes other people hold what might be considered frowned upon opinions. Make respondents feel that the answers they give are socially desirable through use of euphemisms.

Introduce some face-saving phrases, for example 'do you happen to know', rather than 'do you know'. If your question involves criticism, provide an opportunity for praise first before moving to the critics-this way respondents will not feel that the facilitator is unfair by criticizing. Laugh with the respondents' but not at them. Exclaim when they say something intended to be astonishing. Make supportive statements and never argue with the respondent even if you disagree with them.

When interviewing people whose situation is close to one's own you are an 'insider' and in this case participants maybe open up and be honest with you. Be aware of personal reactions and determine whether they fit the research situation or are extraneous. Do not censor information because honest answers from participants are important. Respondents may feel they need to 'explain' themselves, account for their feelings to the interviewer; they may present themselves as victims or powerful knowledgeable storytellers. For instance, when they cry- they position themselves as 'clients' and you may assume a 'therapist role'. Avoid leading questions, words with double meaning, assuming questions and questions about things that the respondents would not know.

 

Conclusions

Focused Group Discussions are a method of collecting data that can be effectively utilised for community-based studies, as engagement is commonly face-to-face and allows for convenient and functional observation. The understanding and application of ethical principles is crucial when undertaking research as this ensures that validity and fairness of the study. The implications of not having consideration for ethical practices will negatively impact a study and possibly jeopardise any future or ongoing opportunities for research. Research ethics are particularly important in conducting FDG, where engagement with the community can be potentially complex and sometimes volatile. Researchers are compelled to ensure that the community in which they intend to conduct research is educated about the study, its importance, what their rights are as participants and other relevant aspects. Researchers are also able to sufficiently prepare for varying power dynamics as they mobilise the community for participation.

There are areas in which power dynamics need to be attended to such as paring co-workers/facilitators, matching facilitators to the group and creating peer groups. Incompetent facilitators are ill-prepared and not well-read facilitators, facilitators who present with a language barrier, facilitators who fail to observe ethical conduct and these are threats to the study validity. This paper reflects on some of the behaviours of facilitators and moments at which these may be altered to achieve research outcomes. This paper also empowers researchers who choose FGD as their method of data collection to normalise similar experiences and encounters.

 

References

Kessi, S, Kaminer, D, Boonzaier, F & Learmonth, D (2019) Photovoice methodologies for social justice. In Laher, S, Fynn, A & Kramer, S (eds) Transforming Research Methods in the Social Sciences (354-374): Wits University Press.         [ Links ]

Kiguwa, P, Nduna, M, Mthombeni, A, Chauke, P, Selebano, N & Dlamini, N (2015) Half of the picture: Interrogating common sense gendered beliefs surrounding sexual harassment practices in higher education. Agenda, 29(3), 1-12. doi:10.1080/10130950.2015.1052678        [ Links ]

Macleod, C & Morison, T (2015) Men's Pathways to Parenthood: Silence and Heterosexual Gendered Norms. South Africa: HSRC Press.         [ Links ]

Macun, I & Posel, D (1998) Focus groups: A South African experience and a methodological reflection. African Sociological Review, 2(1), 114-135.         [ Links ]

Mavungu, E M, Thomson-de Boor, H & Mphaka, K (2013) "So we are ATM fathers": A study of absent fathers in Johannesburg, South Africa. CSDA Research Report compiled by Sonke Gender Justice and the Centre for Social Development in Africa, University of Johannesburg.         [ Links ]

Mnguni, N (2016) Circular Migration and Female Condom use in Hetero-sexual Relationships. The Oriental Anthropologist, 16(2), 417-430.         [ Links ]

Morison, T, Mtshengu, A, Sandfort, T & Reddy, V (2016) "As long as they behave themselves": Heterosexual recuperation in South Africans' talk about homosexuality. PINS (Psychology in society), 51, 28-54.         [ Links ]

Moroke, M & Graham, T (2020) The Position of 'Power' when Engaging Marginalised Youth. PINS (Psychology in society), 60, 41-61.         [ Links ]

Nduna, M, Sikweyiya, Y, Khunou, G, Pambo, V & Mdletshe, T (2016) Vicarious and Secondary Trauma in Interviews on Father Identity and Connections. In Nakray, K, Alston, M & Whittenbury, K (eds) Social Science Research Ethics for a Globalizing World: Interdisciplinary and Cross-Cultural Perspectives (290-303): Routledge.         [ Links ]

Ngunyulu, R N, Mulaudzi, F M, Peu, M D, Khumisi, O & Sethole, M (2016) The Sexual Health Needs of Learners in Makapanstad, Hammanskraal. The Oriental Anthropologist, 16(2), 341-359.         [ Links ]

Oktay, J S (2004) Grounded Theory. In Padgett, D (ed) The Qualitative Research Experience (pp. 23 - 47). Toronto, Canada: Thomson, Brooks/Cole.         [ Links ]

Ramose, M & Baloyi, L (2020) Challenging the teaching of research methodologies in pursuit of epistemic and social justice in South Africa. PINS (Psychology in society), 60, 9-30.         [ Links ]

Spiegel, A D & Mehlwana, A M (1997) Family as a social network: Kinship and sporadic migracy in the Western Cape's Khayelitsha. Co-operative Research Programme on marriage and family life, Human Sciences Research Council.         [ Links ]

 

 

1 https://www.chathamhouse.org/chatham-house-rule
2 Menstruation is commonly referred to as 'periods'
3 http://edition.cnn.com/2008/WORLD/africa/06/24/mandela.quotes/