3. Anthropological Methods -  and the Fieldwork in Quỳnh Lưu

This chapter provides a general introduction to anthropological methods through a discussion of some methodological lessons learnt during our field research in Quỳnh Lưu district.

What are the most important health problems here?  This is an open question of the kind that medical anthropologists often start out with when they initiate research in a new area. There are parts of the world where this question will spur long explanations among local administrators and authorities as well as other people. In Vietnam, however, the immediate answer to this question is that everything is fine, progressing and developing. When we initiated our work in the two communes we were first met with this idealized picture of realities, a picture where everything is the way it should be. This picture is, as we will discuss below, the result of a certain interview context, where people had reasons for telling us what they told us. The following quotes show what picture of the reproductive health situation we were first given:

- Pregnant women do not go on diets during pregnancy. The young ones now have scientific knowledge so they don’t diet. Now they eat the good things: vegetables and fruits and so on.

- Ninety-five percent of women use contraception. They usually use the IUD because there are hardly any side effects with IUDs.

- In this commune, are the wives beaten by the husbands?

- No rarely, because they are afraid of the neighbours ridiculing them.

- Do you have any problems with STDs here?

- We do not find any cases. Here there is no woman who has STDs. There is no prostitution because people fear the public opinion.

- There are no drugs here. There used to be drugs in other hamlets, not in mine. But now there aren’t... And there is no prostitution.

- What about on the beach?

- Yes, there are sexual evils on the beach. But people involved are not from this area. They come from other areas.

- What about premarital sex and abortions?

- There are no premarital abortions.

- So, nobody gets pregnant before marrying?

- Yes, but very little. Only one in 100. There was one man who made a girl pregnant before marrying her, but he came from another place and bought land here.

These statements exemplify the picture that people would like to give us of their commune. It is a picture of a local community where everyone is living in a family and all families are happy. It is a picture where people live in harmony with each other, doing only what they are supposed to do. In this picture, there are no domestic conflicts, no pre- or extramarital sex, no problems with the control of fertility, and everyone is well informed about how to act during pregnancy, delivery, etc. Not only political leaders, chairmen of mass organizations, and health care authorities, but also ordinary people in the two communities often did their best to present an idealized picture of their daily lives, trying to hide their difficulties and troubles. They all seemed to have a strong desire to prove that they are capable of providing a good life for themselves and for their community. There was therefore in many interviews a marked tendency to downplay problems or a reluctance to talk about them at all. This downplaying of local problems in order to give outsiders a positive impression of the community may be a way in which local people demonstrate support of and loyalty to their leaders. But at the same time it may also be a way in which people, leaders as well as others, express ‘how things should be’ rather than ‘how they are’ due to an idea that if one expresses ‘how they should be’, one is already working towards achieving these goals. In addition, many reproductive health problems have to do with private issues that are in themselves difficult and sensitive to talk about. It can therefore be very complicated to get access to people’s experience of their reproductive health problems, whether using a survey or qualitative methods. Much of what we learn is not only in people’s words, but also in their laughter, their silence and in the fears expressed through body language during the interviews. There is nothing unusual or strange in this. The sensitive issues of reproductive health, the high political profile that population and reproductive health issues have in Vietnam, the short period of fieldwork and the high number of researchers involved in this project, however, all contributed to making this particularly present. In this section we will present some of the methods in anthropological research which each in their own way can help us get beyond the official picture - combining the words of people with their laughter and silences and thereby get closer to the insider’s perspective. Before turning to these methods we will present the practical realities of the fieldwork.

The Fieldwork in Quỳnh Lưu

The team consisted of 25 people (including translators and teachers) who stayed in the guesthouse on Quỳnh Bảng beach. The team worked in two groups:  the Community Group and the Health System Group, the idea being that the Community Group focused on people’s experience of their reproductive health needs and problems, and the Health System Group focused on the health care system and health care seeking behaviour. While the Health System Group did interview more practitioners of the various sectors, there was so much overlap between the interviews of the two groups that it was not relevant to keep up the distinction. Most interviews would cover both the experience of reproductive health and the health care seeking behaviour - since, of course, in reality these two things are closely connected, part of the same experience, and therefore talked about together. Analytically, however, we will still keep up the distinction between community experience and health care seeking behaviour when we present the findings in chapters 4 and 5.

One of the two groups would leave the guesthouse in the morning on bicycles or motorbikes to go and interview people in the two communes. The researchers worked in pairs, one asking questions, the other taking notes. Meanwhile the other group would stay in the guesthouse and write and type notes from the previous day’s work. In the afternoon they would go out and interview while the first group would work in the guesthouse. In the second week, the morning group worked in the afternoon and vice versa. Both groups visited both communes in the course of the two weeks.

The teacher from Denmark would discuss and plan the work of the day with each group before they left for the communes, and most days she would accompany one of the pairs and carry out the interview in collaboration with them. Afterwards the three of them discussed the interview, its problems and its outcome. The teacher once or twice in the course of the two weeks, thus, supervised all the participants. In addition all participants met two to three evenings per week to discuss problems and progress of the research.

Two participants in the first week and four in the second week were full time occupied with the collection of existing data on the community and the health care system from the People’s Committees, health workers and other sources. This is the data on which the profiles of the communes (chapter 1) are based. People responsible for this part of the work were all from Nghệ An (Vinh City or Quỳnh Lưu District).

Elaborate plans for the selection of informants had been made beforehand, since the local authorities required this. In reality, however, the choice of informant was often made on the spot by the ‘guide’ of the day. These guides were the leaders of the Women’s Union and the heads of the health centres (or at times other health workers). They selected the informants to be interviewed and they often accompanied the researchers and observed the interview. However, due to the high number of researchers, not everybody was accompanied by local leaders all the time. In addition, as time went by we sometimes approached people independently of our guides who were either busy elsewhere or considered it unnecessary to continue to accompany us.

The groups had question guides covering the topics we had identified as being included in our understanding of reproductive health. These were: adolescent reproductive health, fertility control, abortion, infertility, pregnancy, delivery, maternal health, child care and child health (children under five), STD, RTI, HIV/AIDS, menopause, breast cancer and cervical cancer, sexual violence, domestic violence, gender issues (men’s involvement), education and information on reproductive health issues. Not every interview covered all these topics, but each pair of researchers tried to get around most of the topics in the course of the two weeks. Sometimes the guides attempted to select informants according to one of these topics (i.e. being pregnant, having/having had a RTI, having had an abortion, etc.). This however was often done in ways that appeared insensitive (and unethical) and it was instead recommended by the research team that we attempted to identify people in the reproductive age without the health workers and local leaders publicizing their knowledge about people’s intimate problems. In summary, the selection of informants did not take place in a systematic manner and was not always controlled by the researchers. Still, we managed to interact with a broad range of people. Considering the general, though sensitive, character of the topics we wished to interview people about, all informants had important contributions to make to our research. Whether we managed to reach a broad enough range of people (including the most marginalized) is difficult for us to judge. However, we clearly did not talk only to privileged members of the community.

In the ongoing evaluation of the research while still in the field we realized that some topics were only sporadically or superficially covered in the interviews: e.g. sexually transmitted diseases, sexual violence, non-marital sexuality, child care and child nutrition, while others were recurring frequently: infertility, contraception, pregnancy/delivery, RTIs etc. There may be many reasons for this: the focus of the interviewers, the prevalence of the problems, the experienced severity of the problems, and their sensitivity.

In total 72 interviews were carried out (only three persons were visited more than once). In some cases both the man and the woman of a couple were interviewed either together (once) or, as was usually the case, on separate occasions. Twelve of the 72 interviews were with local leaders (political leaders, chairmen of mass organizations) and nine were with health workers (from family planning volunteers to the heads of the health centres). Seven were health practitioners from the folk sector (prayers, fortune tellers, practitioners of oriental medicine as well as untrained pharmacists). Only two unmarried women and one unmarried man were interviewed and only one of these was below 20 years old. Thus, we have very limited information about adolescents. With few exceptions all other persons interviewed were in their reproductive age though a couple of the women had reached menopause.

Research Questions

Research questions and objectives in qualitative research are usually very broad and prone to ongoing adjustments as the research progresses and the researchers learn what is important in the area where they carry out the research. The research questions for the study on which the present  publication is based were

1)   to study people’s experience of their reproductive health problems and needs, and

2)   to describe the health care options available in the area and people’s health care seeking behaviour in relation to reproductive health.

In chapter 2 we introduced some of the core concepts of medical anthropology used when studying people’s experience of their body and health and when studying the health care system and health care seeking behaviour. In this chapter we will discuss the methods used to obtain these insights. But first a few comments on how research questions are adapted and shaped in the course of the research.

Based on these open research questions we aimed at getting background information which together with the survey results would provide the basis for future more focused in-depth studies.  But  already  in  this  first  small  study  we adapted  the research questions as the research progressed. We started focusing more on family life and problems like infertility, divorce, single parenthood and other issues related to being unsuccessful in obtaining the ideal and happy family that everybody hopes for. We ‘narrowed’ the research question, realizing  that we could  not  during  this short  fieldwork  obtain

information about all aspects of reproductive health. At the same time we may say that we ‘broadened’ it in the sense that we opened up to a part of the context - social relations, family life - which we had not initially given priority to. These adaptations were shaped by several things, two of which are: a) the above described problems of doing fieldwork on intimate issues related to reproductive health made it difficult for us to get in-depth information about e.g. sexuality, STDs and RTIs, and b) people with problems of infertility, divorce, etc. were often more talkative and less pre-occupied with the official picture, maybe due to the fact that it was obvious to everybody that their lives did not fit into this picture.

Adapting research questions by ‘narrowing’ them in certain ways and ‘broadening’ them in other ways according to the empirical reality met in the field is an inherent part of qualitative research. As will be discussed later, analysis is part of all stages of the research and takes place already during the collection of data. The ongoing adaptation of the research questions is one example of this.

Basic Principles of Qualitative Research

Anthropology as a discipline is closely associated with qualitative methods. Today many anthropologists also use quantitative methods, and qualitative methods are used by other disciplines than anthropology. Still, anthropology is largely defined as a discipline that employs qualitative methods. This is the approach we will be presenting here.

The methods used in medical anthropology do not differ from methods used in other branches of anthropology. In all kinds of anthropology, the overall aim is to try to understand local worlds from the perspective of those who live in them. Anthropologists use a range of different qualitative methods to do this, including observations, interviews and conversations, case stories, focus group discussions, and analysis of documents. We can only discuss a few of these there. Before briefly considering the three most important of these methods – participant observation, interviews and case stories – we shall have a look at some general methodological principles in anthropology.

Spending Time in the Field

Generally, anthropologists try to spend as much time in the field as possible: if we want to really understand how people live their daily lives, it is not enough to pay a quick visit and ask how things are going. We need to spend extended time in the field, having time to observe life in the local community on a daily basis and get to know people well. The need to get to know people well is particularly urgent if the research concerns issues that are personal and sensitive, as is the case in reproductive health research. In order to learn about intimate aspects of people’s lives, we have to establish relations of trust and confidence first. In order to complement the more public and factual information that can be gained from structured interviews or surveys, it is therefore important for the anthropologist to get to know people in a way so that they feel comfortable talking about their more private feelings and experiences. This usually requires time. It also requires tact and respect for other people’s ways of doing things and living their lives.

Meeting Local People with Respect and Humility

When doing field research, anthropologists usually try to meet local people with humility and respect. There is a certain relativism to the anthropological endeavour in the sense that we try not to judge other people, but instead to understand why they act as they do – why it makes sense to them to do as they do. For instance, when observing a woman coming to the health centre to have her fourth abortion, it would be easy to simply condemn her for her careless sexual relations and her lack of ability to prevent pregnancies. Instead of simple condemnation, however, the anthropologist would always try to meet her with openness and understanding. We can talk to her about her experiences with contraception, about her sexual relations to her husband, and about her relations to local cadres and health workers. We can try to understand her feelings about having to undergo another abortion and the constraints she faces when trying to control her sexual and reproductive life. If we try to understand her instead of making moral judgements on her situation, we will perhaps learn enough to be able to make suggestions for how policies and programmes can better meet the needs of women like her. A core assumption in anthropology is that local people are not ignorant, even though policy-makers and planners may find that they lack knowledge. They always know something – it may not be what researchers or authorities think they should know, but they do possess knowledge. For instance, the abortion-seeking woman mentioned above may seem not to know how to prevent a pregnancy. But the knowledge she does possess is that modern contraception can be harmful to her health and that it is important for a woman to satisfy her husband sexually – and this is the knowledge that she acts upon. To enable the health care system to meet her needs, we have to recognize her knowledge, her perspective, instead of simply condemning her as ignorant.

We will now turn to some of the most common methods used in qualitative research: participant observation, open interviews and case-stories.

Participant Observation

The primary research method in anthropology is participant observation. Participant observation consists of two different activities: participation and observation. Through participant observation we try to take part in local lives as fully as possible, seeking to share the experiences of people we study. We try to participate in life, at the same time as we observe it. An anthropologist studying children and nutrition, for instance, will not only conduct a survey on food intake or talk to mothers about how they feed their children. The anthropologist will also try to spend as much time with families as possible, going to the market, preparing food and having meals together with local families. In order to obtain a more full understanding of the wider context of daily lives of which food and nutrition is a part, the anthropologist may also take part in other activities locally, such as working in the fields, washing clothes, watching television, attending weddings and other celebrations, etc.  Through these activities, it is possible to learn more about what is important to people, about the constraints they face in their daily lives and about their motives for acting as they do. For instance, the anthropologist studying child nutrition might learn that in most families it is mothers that are responsible for feeding children. Yet mothers are also very busy with a multitude of tasks and do not have time to ensure that children actually have enough to eat at meals, hence many children are underweight.

Participant observation can take many different forms. On a continuum of involvement, we can distinguish between complete, active, moderate and passive participation (Spradley 1979). In passive participation the anthropologist is present, but merely as an observer who does not actually take part in activities. This is often the case in studies conducted at health facilities, where interactions between patients and health staff are observed. Active participation, in contrast, involves trying to do what other people are doing, in order to learn more about their experiences and motives. This is often the case in community studies where the anthropologist lives with local people for an extended period of time. Whether we live with people for extended periods of time or visit their area for a short period, participant observation takes place all the time, even during an interview. We observe what happens on our way to interview people. While we talk to people their bodily expressions may give us information that is not expressed in the words they use. They may tell us something important when they escort us as we leave. We observe while going to the market to buy food for ourselves, or while waiting for the bus or the man who repairs our bike. The researcher is never completely ‘off duty’ when on fieldwork. What happens in between interviews is as important as what happens during interviews.

Participant Observation in the Field

Ideally speaking anthropologists do as mentioned live for an extended period of time not only among but also together with people they study. They should, to the extent it is possible, become part of everyday life, participate in the same activities as people who live there, and learn through their participation. In reality it is rarely possible to fully do that. It should, however, still be attempted to approach the lives of people not only through interviews, but also through simple ‘presence’ in their lives. This proved extremely difficult during the short fieldwork we carried out, for several reasons.

Firstly, there were too many of us. Twenty-five persons will always remain a disruption in a small community (though maybe an entertaining one). Secondly, people in these two rural communes were not used to having strangers around: neither a large group of people from Vinh and Hà Nội, nor Westerners. We were a curiosity and in many people’s minds a potential threat to the local order. In particular the presence of the foreigner (the Danish teacher) caused concern as well as curiosity. Thirdly, anthropological research is a new and foreign concept that nobody, leaders or others, we met in the field had had any experience with or had heard of. It was therefore difficult to explain convincingly that we were not there to check on, evaluate or report on local people or local authorities, but to learn about daily lives and experiences.

No fieldwork situation is ideal, and it is always a challenge for researchers to find out how they can gain some access to people’s everyday lives. We will give a few examples of how ‘participant observation’ was a part of our fieldwork, even if only in limited ways. These examples will illustrate the continuum of involvement mentioned earlier, where we can distinguish between passive, moderate, active or complete involvement.

Example 1: We planned some systematic observation. A couple of researchers were hanging out in the health centres and the drug stores to observe the interactions taking place there. One member of the team wrote the following as part of his observations at the commune health centre:

The plot of land on which the health centre is placed is about 3 ha large and contains both the commune health centre and the polyclinic. There are 6 buildings, only 5 are in use, but 4 of them seriously need to be repaired. Only the commune health centre, recently restored, is in a decent condition. It was 2 pm. Some rooms were crowded with people: the conference room, examination

room treatment room, but especially the maternity health room was full of people. People were waiting inside as well as outside. When I came closer I saw that there were 5-6 persons in the room. Two, one old and one young, were wearing white blouses and were advising the woman whose abdomen looked bigger than her body. She grimaced a little. She might have been uncomfortable since she was going to give birth. When she saw me her face became more serious. She tried to pretend she had nothing wrong with her, was not in pain. The two nurses continued to console the woman, greeted me and left the room to come and talk to me. We already knew each other, since I have been coming to the health centre every day with the research team to get the help of the health workers to find people to interview. One of them smiled at me and asked: ‘So what have you come for today?’ I replied that I was waiting for my colleagues who were coming. I lied to her to make her more comfortable that I was around than if I told her that I was here to observe them. I left them and started observing in another room. While I was there observing these people, two nurses came and asked me again why my colleagues did not come. I just smiled at them and told them that they would be coming. I tried to avoid talking to them, because I wanted to be able to be around and observe what they say and do without them thinking that was what I was there for. I went and sat in a corner at a distance from where they were working, and they continued their work without worrying about me...

As discussed above we may think of observation and participation as two ends on a continuum. We try to do both at the same time, but there are times where we observe more than we participate or participate more than we observe. In the above example the researcher observed more than he participated in life at the health centre: he was not a patient or a relative of a patient, but he was not a health worker either with any responsibilities there. However, he was not a fly on the wall or observing through binoculars from a distance. He physically entered the life at the health centre, greeted and talked to people, though in that particular situation there was more focus on the observation than on the participation. The researcher decided not to inform the people present that he was actively observing them, because he felt it would make his presence less disturbing. In general, research ethics require that we inform people that they are part of a research process, and it can be problematic to observe and inquire people about information without making it explicit that it will be used for research. In the above situation, however, people at the health centre were well aware that the researcher was part of a team carrying out research in the commune and had been assisting them do so an various occasions. And whether you purposely decide to spend some hours in a certain place or just happen to have to spend time there, you should when doing anthropological research always keep ears and eyes open and note (afterwards) what you learned from your presence there.

Example 2: Participant observation takes place even as part of interviews. This kind of participation may be referred to as moderate. The researchers do not carry out daily activities with the informant, but s/he does more than observe from a distance. S/he actively interacts with the person and learns from that interaction.  The following are two examples of notes of observation that help conceptualizing and therefore analyzing interviews:

Some of the members of the research team had met Mai, a 30-year-old woman, a few days prior to the day she was interviewed. She had been smiling and laughing when meeting them on the small path outside her house and had invited them to come to the house to visit her family. She lived in her parents’ house with her three children and her grandmother. A couple of days later members of the research team visited the family again, this time with the explicit purpose to interview her. Again she appeared smiling and laughing and the interviewers were hoping for a detailed and openhearted talk with this friendly woman. ‘We are a poor family, but rich in heart’, she said several times before she had even sat down to greet the visitors  - and each time she said it the sentence was accompanied by laughter. As the interview progressed and her laughter increased it became clear that this was not a simple laughter of happiness. Whenever her economic situation was touched upon she either laughingly repeated ‘poor in wealth, but rich in heart’, or she sadly and timidly looked down and avoided answering the question. The same happened when we asked her about her 3 children, her deliveries at home, her use of contraception, her use of the health centre.

In Mai’s case, her laughter and silences taught us about the silent difficulties of her life – the poverty she is struggling with and her feelings of shame about being poor. Mai’s choice to sometimes remain silent tells us important things about life in the villages in Quỳnh Lưu. Many people are poor and know they cannot hide their poverty. But poverty is embarrassing to them, and they will do anything they can to make it seem less severe than it is. We learned more about this from observing her interaction with us than from her words. The following is another example from the field diary of one of the researchers:

When we interviewed Hạnh today we felt the atmosphere was very open hearted, more than is usually the case. She is a student at Vinh University who is on summer holidays at home. Even though the foreigner (the teacher from Denmark) was also present she talked very freely. Most people become more afraid when she is there, but not Hạnh it seemed. She talked straightforwardly about the bad attitudes of the doctors and health workers, about her trust in healers using oriental medicine who cured her sister who was seriously sick.

This kind of contextual information can become important for analyzing the interview carried out with Hạnh: we cannot assume that everybody will be as critical about the health care system as she is, since she has been outside the commune, has seen other things. But we cannot either assume that if others do not say anything about the bad behaviours of health workers, it is because this does not take place. Maybe they fear talking about village-mates (the local health workers) whom they may fear will get access to the information they give us. We need to ask again ‘Who says what to whom and why?’ and then compare the different interviews where people told us different things for different purposes, and we need notes on the context and on observations during the interview to be able to do that.

Example 3: Another example of moderate participation shows how it is often not the planned interviews that provide the most important insights. Sometimes informal interactions and spontaneous conversations that turn into more structured interviews can be far more important, hence the importance of participant observation, of noticing everything that happens in between the interviews:

One day three members of the team had been introduced to a woman who  had just  been through an operation. She had returned to her own house, but was still weak, and confused about why we turned up there to interview her. It was still painful for her to sit up and to talk about the operation. We soon realized it was inappropriate to carry out an interview with her and left early. We crossed the road to get something to drink in a small shop by the road. We had no intention but to relax a bit before the return to the guesthouse, but eventually we started an informal chat with the friendly shopkeeper. We returned to see her twice more, and each time we combined friendly interaction with more or less formal questioning and note taking.

The result was the case about Lý presented in the introduction. Again and again, anthropologists have to acknowledge that often the most important information does not arise from planned and formal interview settings, but from strolls down the road, visits to bars along the beach, relaxing informal visits to families where the notebook is not pulled out of the bag at all.

Example 4: The final example will show how two of the researchers moved towards active or almost complete participation as they became involved in the story of a woman with severe difficulties in her life. The leader of the Women’s Union took two of the members of the research team to see Trang, a woman with her own shop on the main street in Quỳnh Bảng. She was chosen due to her recent experience with a stillbirth of an 8-months-old foetus (see case in chapter 4). The 38-year-old woman talked non-stop from we first arrived, and seemed in no way intimidated by the presence of the researchers and the leader of the Women’s Union. Due to her openness the researchers decided to return to her later. In total she was visited three times by four different researchers. Trang’s story and misery moved the researchers considerably and they started feeling guilty for being given her story and have nothing to give her in return. They felt obliged to find a way to assist her in returning to Hà Nội to get help for her infertility. As their relationship to her developed they did, however, also start feeling uncertain about the many contradictions in her story, what she was actually telling them and whether she was trying to ‘exploit’ them by presenting her life as miserably as she did. There is no answer as to whether she was lying or exploiting the researchers. We will never find out whether she was beaten as much as she claims to have been, and why she chose to return to Quỳnh Bảng and whether she was actually a prostitute herself. It is a fact of life that the researchers ‘use’ informants for information and that they will often experience that informants try to ‘use’ them for something in return. This is understandable and acceptable. In many situations anthropologists become involved in other people’s lives during the fieldwork, but they should never forget to use the experience of being involved to observe and learn new things. Complete participation where we no longer observe is not the goal of participant observation.

Interviews and Conversations

Interviews and conversations are some of the most important anthropological research methods. In order to understand how people perceive and experience things, we often have to ask them questions about their lives and experiences. An important difference, however, between questions asked in quantitative research and in anthropological research, is that anthropological interviews are most often open-ended and flexible. In anthropology we try to make interviews resemble ordinary conversations as much as possible. Instead of having a pre-defined list of questions, we often conduct interviews using simply a list of topics to be explored, allowing the informant to ask questions back, to take up new topics, and to take the conversation in unexpected directions. Again, the idea is to try to stay as open as possible to the informant’s point of view and to try to gain insight into the things he/she finds important. For the anthropologist, conducting an interview therefore often means trying to strike a balance between the research agenda and the informant’s agenda – staying open to what the informant finds is important to talk about, while also keeping the interview focused on the initial research questions (for further details see Spradley 1979).

There are no specific guidelines on how many interviews one should conduct in a qualitative study. The only rule is that a small number of in-depth interviews are far better than a large number of superficial interviews during which a relationship of trust between informant and interviewer was never established. It is rarely possible for an interviewer to do more than one or two, maximum three good in-depth interviews in one day. In theory you can do a fieldwork without any interviews, doing only participant observation and  having  friendly  conversations with people whom you come across during the period of fieldwork. In practice, however, most fieldworks include a number of interviews. One should consider beforehand whom it is relevant to talk to and how many it is realistic to interview if one is to also have time to do participant observation and other methods. Again, however, one may have to adapt the number and kind of people one interviews, and again this is okay and part of the research process. If we had planned a fieldwork focusing on women’s experience of contraception and discover that women never make decisions about contraception without consulting their husband, then we will have to adapt our research strategy and start including men. If we decide to focus more on ‘unsuccessful’ families (single, divorced women, etc.) we may have to look for more people living under such circumstances than we had first planned to do. Like with the research questions and the interview questions, the number and kind of informants are also adapted in the ongoing process of doing anthropological fieldwork.

Interviews and Conversations in the Field

A member of the People’s Committee introduced two researchers to a woman. They introduced the interview as follows:

-  Thank you for agreeing to talk to us. We are interested in hearing something about how you care for your own health and that of your small child.

-  I don’t know what to say. Near here there are some women who have also just given birth and who know more than me about these things. You can ask them. Also I am busy.

-  We just want some small information, you can tell us anything you think. Do not feel shy. Tell us about your own reproductive health.

Many people do not understand why they are selected for an interview. They expect the researchers to be interested in professional knowledge. They fear they do not have sufficient understanding or the right information. If they understand that it is their personal experience we are interested in, they may simply be reluctant to share this with us. And why should they? Why should they trust that 25 intrusive outsiders would keep private sensitive information about their lives confidential? Why should they tell us anything, which if brought to the knowledge of the wrong people may intimidate them and have negative consequences for their lives? In spite of this, many people did open up to us, and what we learned from them is presented in chapters 4 and 5. Here we will, however, use the example of a problematic interview, which illustrates some of the things that can be barriers to the development of a relationship of trust, which is necessary for a successful qualitative interview.

One of the interviews was carried out by three of the researchers with a 30-year-old tailor with two children, pregnant with her third one. The interview took place in the front room of her house, which was the bedroom, the living room as well as her workshop. When the interview started, the woman, the three researchers, and one of the health workers from the commune health centre were present, but soon the husband joined, customers entered and sat down to listen, children of all ages were peeping in through windows and door ways. The woman became increasingly uncomfortable about the whole situation, but continued to answer politely to the questions posed to her. Part of the interview went as follows:

-            Why are you pregnant now since you already have two children?

-            I could not use an IUD because it made me feel very tired. I tried twice but each time I had to have it removed.

-            What were the problems?

-            I had my period all month.

-            Did you go to the health centre?

-            Yes, but they could not help me.

-            Did you use herbs?

-            Yes, but no change.

-            Why did you not use another method?

-            We tried to use the condom. But we failed and I got pregnant.

-            Why did you not use pills?

-            I was afraid the pills would make me sick.

-            Did you wish for this pregnancy?

-            No, I did not want this third pregnancy.

-            Why did you not choose to go for an abortion?

-            I have already had an abortion. Three times. The first time was when the first child was one year old. The second time when the second child had just been born. And one more time since then. But this time I have decided to keep the child. I am two months pregnant now.

-       Why, when you already have two children do you still decide to give birth?

-       Because I am afraid of having another abortion. Each time it has made me very sick for very long. I am afraid what will happen if I do it one more time.

At some stage the woman seemed so uncomfortable about the situation that one of the researchers decided to interrupt by asking questions about her work as a tailor, the patterns she uses for sewing and the material she sells in her shop, and thereby ending the interview.

The above interview is an example of a problematic interview, which in many ways does not live up to the criteria of an open interview, but which nevertheless, with the help of participant observation, does teach us about life and reproductive health in Quỳnh Lưu and which also points to some important issues concerning how to carry out open interviews:

·      The woman felt intimidated right from the beginning due to the presence of so many people: two researchers from Hà Nội, one foreigner, and one representative of the local authorities. The curiosity of customers, and her own and neighbouring children only worsened this. Whether her husband’s presence was a reassuring factor in this situation or further added to the discomfort of talking about reproductive health issues is difficult to know. But it is certain that in order to create the intimacy and openness that makes qualitative interviews useful, the researchers often need to be alone with the informant.

·      The approach of the researchers did not improve the situation. The way the questions were asked made the interview seem like an interrogation: ‘Why did you not do what was expected of you?’ The only option the informant had was to answer in defensive ways that do not necessarily give much insight into her own experience of the problems she has been through.

·      The researchers did not have the ‘insider’ as a starting point. They had the outsider’s point of view. They had as a starting point the discourse of the family planning programme of Vietnam: each woman should have only two children, she should use an IUD or another contraceptive method to ensure this, and if an accident occurs she should have an abortion. Their interview resembled an attempt to inform and convince the woman of this policy, i.e. to promote an outsider’s point of view. Irrespective of the benefits or problems of the family planning policy, an open interview should strive towards eliciting the insider’s point of view, not to convince the informant about the outsider’s point of view. What we wanted to know from this woman was the problems and pains she went through, how did she experience her reproductive life, how did the national family planning policy influence her life, her happiness and her troubles, what options did she have to act and to make decisions, who supported her, etc.?

·      The interviewers did not probe. After one short answer to their question, they went on to the next topic. Probing would indeed have been difficult because the woman was already so uncomfortable. But the result is that the interview does not provide any quantifiable information, but it does not provide much in-depth information either. The fairly closed questions and the lack of probing makes it fall somewhere between open and closed interviews, and neither one nor the other.

·      The information we get is always dependent on context. No matter how good and experienced a researcher you may be, situations like the above occasionally arise and the researcher can do little or nothing about it, except maybe end the interview and still learn what there is to learn from it by always asking her/himself: ‘Who said what to whom and why?’  There may have been reasons for the problems with condoms and abortions that the woman did not wish to talk about in this situation. She gives us answers that she would give to the family planning co-ordinator, leader of the Women’s Union or leader of the People’s Committee, had they asked. She explains how she did what she was expected to, but there were faults in the system (IUD gave side-effects, condoms were insufficient, abortions a danger to her health). She tried, the way the people listening to the interview would like her to have tried, and the failure was not her fault.

·      Participant observation: It was obvious to everybody that the woman was uncomfortable and that this influenced the interview. But observations also came to play another role in the analysis of the interview. As one of the interviewers said afterwards: I understand that she does not feel comfortable having another abortion. I have seen the facilities in which these abortions take place and I would not feel comfortable and safe about my health if it were me. Observations made by the researchers during as well as prior to the interview often become of importance for analyzing the information that the interview provides.

·      In spite of the intimidation and the informant’s fear, the simple account of a reproductive life story presented here does give us insight into how difficult it can be, in spite of the efforts and services offered to people, and in spite of people’s intentions to make use of these, to actually realize the goal of the family planning programme.

In a sense, this short interview summarizes some of the main themes in people’s experience of their reproductive health, themes that recurred in many interviews and observations. But qualitative research is not of much use unless we manage to create situations where we get deeper into these topics and get more information on how they are linked to other aspects of the everyday lives of people and of their relationships to each other. To do so, the style and tone of the interviewers must be different, the setting and presence of other people must change, and the woman must develop a relationship of trust with the researchers. This is a lot to ask from the short fieldwork and first time experience of the participants in this research, but in chapter 4 and 5, however, we will present findings which show that it was indeed possible to get more in-depth information even within the limited time and possibilities of this fieldwork.

Case Studies

Case studies consist in detailed descriptions of local events . It can be the detailed description of one person’s life, or of one person’s involvement in a certain event. Anthropologists favour case studies because they allow for descriptions of people, things and events in their context. It is the detail and particularity that distinguishes case studies from other anthropological accounts; case studies describe in great detail how specific events or lives unfold over time. The case of Lý introducing this publication is an example of how one life as it unfolds over time can be presented. While the case of Lý focuses on issues of relevance to her reproductive life, it also covers contextual factors that shape this life. It is, again, an example of how the qualitative researcher has to constantly find a balance between the research agenda and the individual’s agenda. Not everything in Lý’s life will be interesting for us to hear about, but at the same time we have to be open to what she sees as important in shaping her reproductive life. We only met Lý three times and interviewed her twice, but a case-story may be composed of many more interviews than that and participant observation over an extended period of time during which the researcher comes to know the person or family being studied.

A case study can be a study of an individual, a family, a section of a community or a whole society, but in anthropology it is often an individual or a family.  One of the common objections which researchers from other disciplines have about the anthropological use of case stories is that cases focusing on individuals or families cannot be representative, i.e. cannot be used to say something general about the society or situation being studied. It is important to keep in mind that we do not study one person or one event for the sake of that person or event, but in order to understand how they are connected to the whole - to other persons, to other events, to health problems, to economic and political situations, to religion, etc. By studying these connections in detail we may discover problems we could not foresee and therefore have asked about in a survey. What we learn from a case may not be representative, but it has a higher validity than answers in a survey, because we know people outside the interview context also. By studying a person in her context we get insight into ‘What she says or does to whom and why?’ We start understanding her reasons for acting the way she does - but not how many people act like her or how often she acts like that. We cannot with the case study method (or other qualitative methods) find answers to the question about how many refuse to turn up for gynaecological check-ups, but we can get insight into some of the reasons why they do not turn up - reasons that are also valid in other contexts than the specific context represented in the case story. The case study method may also help us to get insight into discrepancies between the figures stated in statistics and real life events. For instance, in a social setting where low levels of maternal mortality are officially reported, the case study method may reveal that actual numbers of deaths are higher than what is revealed through statistics (Bale 1999).

Recording Data

The quality of an anthropological study is highly dependent upon the quality of the notes taken in the field (cf. Sanjek 1990). Since participation and observation are important methodological tools, the ‘results’ of qualitative research are not just the words spoken through an interview. The notes taken in the course of the research should also include the daily observations and experiences of the researchers. When taking notes from an interview, the researchers should include observations about the setting of the interview, the atmosphere, and the information communicated by the informant through other than verbal means. The researchers should also keep a diary where they take notes of their observations about life in general in the local community and about the events and things they come across during the day. When the research is analyzed and written up, these observations may become important. They are important in themselves, providing factual information about lives in the area, but they are also important as a means to conceptualise and get a deeper understanding of the verbal statements made by people in the field.

The exact words and lengthy descriptions of the informants are often of interest to the research process. It is important not to summarize in the researcher’s own words what the informant said. If for instance we summarize and ‘translate’ the explanatory models and illness terms people use to diseases and diagnoses we know from Western medicine we may miss important information (cf. chapter 4). But not only people’s own terms should be written down, sometimes also whole expressions and sentences may carry information that is lost if we simply summarize the main points of what we ‘think’ they are trying to tell us. Therefore, use of a tape recorder can be an important support in the recording of data. There are certain limitations, however, that we need to be aware of. First, the tape recorder sometimes makes people feel shy and uncomfortable. People may feel uncertain about how their words are going to be used by the researchers – and if so, use of a tape recorder may heighten their anxiety. Second, it takes a very long time to transcribe an interview. For each hour recorded it may take six to seven hours to transcribe the interview. In other words, there are arguments for and against the use of tape recorder. It must be considered in each situation whether it is appropriate to use it or not. Therefore, note taking during and after interviews, and the writing of diaries remain the most important techniques for recording qualitative data.

Taking Notes in the Field

It was soon decided not to use tape recorders in Quỳnh Lưu due to the limitations discussed above: people were already very intimidated by our presence, and it would have been extremely time consuming to transcribe the interviews. In addition the participants worked in pairs, and one person concentrated on taking notes. They made the note taking very extensive and detailed. In cases where researchers work alone it is difficult to carry out as extensive note taking. A tape recorder was used on a few occasions. The first time we met somebody, we would usually not suggest turning on a tape recorder. But particularly talkative persons who willingly welcomed us a second time were asked if they would agree to have the interview recorded.

Two members of the team carried out observations in the commune health centre and pharmacies in the course of two afternoons and noted their observations. Everybody was encouraged to write diaries, and five researchers did so. Notes from many of the interviews also include comments and observations made during the interview.

Thirty-eight interviews and three diaries were translated into English. About two thirds of these were translated already during the fieldwork, so that the Danish teacher had an opportunity to follow the process and comment on the interviews. A second Danish researcher who knows Vietnamese took part in the analysis after the fieldwork and training course had been completed.

Analyzing Data

In quantitative research, data is first collected and then analyzed according to fairly formalised principles. After this the results can be presented and discussed. In qualitative research analysis is a continuous process, which begins while data is being collected. The analysis starts already while interviewing, since questions are continuously developed and adapted to what has already been told. Which interviews to carry out, which topics to cover, and who to interview are also decided during the process of research where the researcher continuously learns from already conducted interviews. As we start writing, the presentation of data is combined with a discussion of these. In qualitative writing we cannot make the kind of separation into results, analysis and discussion, which is common in quantitative research. It is an integrated process (cf. figure 1). But how does one get to this integrated text?

Notes from interviews, and even from observations and diaries, tend to be fragmented and incoherent. Informants often jump from topic to topic. The notes may contain a lot of information that is not useful, or at least not useful for the researcher’s present research purposes. We therefore have to start by systematising the information. We have to order and reduce the data. Ordering is best done in relation to the research questions and topics. Ordering is done by deciding on a coding system, where a code is written in the margin of the text. Several codes may be assigned to a certain passage or sentence. It may be necessary to review the data several times, doing preliminary or open coding, which is later elaborated on with codes referring to the discussion of topics in the written account of the research. This coding, however, is but the very first step in the analysis. The codes are not the results. They simply help us organize the data and identify the topics to discuss in the final text.

Once these topics are identified we start the process of comparison (cf. chapter 2 on analytical concepts). If one topic is infertility, we start comparing what the same woman has said about infertility in different parts of the interview, what different women say about infertility, what men and women say, what we know from other parts of Vietnam, from other parts of the world etc. Doing these comparisons will help us reach an in-depth understanding of the issue of infertility and its consequences for those who live with it.

Analyzing Data During and After Fieldwork

During this fieldwork, as always in qualitative research, we started the analysis of data while still in the field. Every day, when interviews or observations had been done, the researchers read through their notes and transcripts, reflected upon what they had learnt and developed new questions for further field research. The coding of the collected data was done during a data analysis workshop held in Hà Nội after the fieldwork had been finalized. Coding was done in several different ways, using:

1. The topics that the researchers had defined as part of reproductive health (e.g. contraception, abortion, pregnancy, delivery, etc.)

2. The research questions: the health care system, and the local experience of reproductive health, and the analytical concepts chosen to explore these (health care seeking behaviour, illness terms, illness experience, interaction with health workers, folk sector, professional sector, etc.),

3. Issues that grew out of the material but were not part of the research questions or the conceptual framework from the beginning (poverty, divorce).

As an example of the process of data analysis, we shall now return to the case of Lý who introduced this book. This case was not told to us in the exact words we presented it in the introduction. The way it is presented is a result of several rounds of analysis, combination of interview data, participant observation, use of the research questions to remind us what we are interested in, use of anthropological analytical concepts, topics that appeared in the course of the fieldwork, which are also relevant for that case, and knowledge of what the continuation of the text will be, i.e. what interests us in the chapters that follow. The case of Lý is presented as the very first thing, but it was not the first thing to be written, and it has been rewritten several times. Not everything we know about Lý or about the reproductive health problems people are faced with can fit into a case covering only a couple of pages. What exactly we find relevant to include depends on the process of analysis and writing. This is a circular process, which we will now try to illustrate.

Below is an example of how the interview with Lý can be coded with a) topics from the reproductive health list, b) analytical concepts from research questions, c) a later understanding of what is relevant for the research.

As this illustrates, the case you read in the introduction is a result of interviews as well as participant observation and a long process of analysis over various stages: during the interviews, ordered according to the topics of reproductive health, ordered according to the research questions and ordered according to the topics appearing in the course of this particular process of research. The case was presented with certain purposes: the purpose of showing how reproductive health interlinks with other aspects of life and the purpose of showing how people sometimes strive hard to create a ‘happy family’ – and yet fail.

In the next chapters we will present some more of our research results. When reading these chapters, it is important to remember that this text has been through the same circular process of analyzing over various stages. These stages are only at times visible in the final account. In other words, the text is a choice of representation, which we consider useful for approaching new understandings about reproductive health on the North Central coast and in Vietnam more generally.

Previous Chapter                Next Chapter