Contents - Previous - Next

3. Training the study team

Sensitizing the study team
Transferring technical know-how
Management, review, and analysis of information
Developing working hypotheses
Outlining the study design
What resources are needed?
How much time should be allowed for training?

Once you have established your study team, you will need to assign a suitably experienced applied anthropologist or related social scientist, a trainer who can guide and supervise the study team during planning, designing, and conducting the study. Some projects already have highly trained staff. Training may be given periodically by external or in-house trainer(s). Other projects may have less trained and/or skilled personnel. This chapter provides some guidelines for the trainer who must take into account existing skills when planning and conducting initial as well as on-the-job training.

The purpose of training is threefold:

• to prepare and sensitize the study team for the disciplined task of qualitative investigation of the sociocultural determinants of hygiene practices;

• to equip the team with the technical knowledge and skills required to carry out such an investigation;

• to involve the study team in designing the study with a view to conducting it themselves.

These three are equally important purposes of training because the attitude and behaviour of the study team can influence the quality of information obtained, as much as the team's skills of investigation and analysis can. In addition, it is easier for the study team to conduct a study designed with the help of team participation than it is to carry out a study that has been designed by one or more individuals prior to the team's engagement.

Sensitizing the study team

To sensitize means "to make sensitive" and is a term widely used in participatory research as well as in anthropological circles where the ability to recognize local/lay knowledge is highly appreciated. The term is less known in circles that uphold the premise that knowledge comes from the academic and/or technical "expert" and should be transmitted to the "ignorant lay or local person.

Investigating hygiene practices systematically is not something that comes naturally to the average water supply and sanitation project worker, particularly if his/her training background and/or experience does not include qualitative investigations. For example, the average water or sanitary technician concentrates on pumps and pipes and does not see the need to find out what the users of those pumps and pipes are doing; what they have to say, and why.

It is important for each member of the study team to be trained to be inquisitive and not assume that she/he knows what is going on in the locality. It is very important to adopt a learner's attitude, however difficult that might first appear to someone whose job it has been to teach the people.

Encourage the study team to discuss the following ways of establishing rapport with the study population, and to identify the most relevant points for their own setting. People will react towards members of the study team according to their gender, ethnicity, age, style of dress and speech, and how they present themselves. Individually and as a team, investigators should aim to minimize status differences so that people can feel comfortable with them. This will bring about cooperation and minimize the "observer effect." where people act or say things that are not usual because of the presence of strangers. It will also help to minimize possible bias as a result of the observer effect.

Suggested ways of establishing rapport include:

• Appearance. This is very important. For example, when working in a rural community, try to identify with local people and minimize the gaps that exist by wearing simple clothes and using simple language. The reverse may be necessary when talking to professionals or top level personnel, you would dress to suit the occasion, but without drawing unnecessary attention to yourself.

• Greeting. This is more effective in the local language, especially when in rural settings. If you are not a speaker of the native language, it is helpful to learn at least a few basic words to demonstrate your interest.

• Introduction. Introduce yourself and ask the person you are addressing to introduce herself or himself, in a locally acceptable manner. This is particularly important in the case of interviews. In group discussions, each participant should be invited to introduce themselves. This will help to assure study participants that investigators are genuinely interested in learning about them.

• Terms of address. When asking questions, use the respondent's name, with the appropriate title, whenever applicable. This helps information gathering (particularly interviews) to remain informal or conversational rather than formal or interrogative. The person being asked questions should not feel that they are on trial or being given a test about what they know.

• Establish confidence by stressing to the study participants that you are interested in her/his/their opinions, knowledge and beliefs. In the context of individual interviews or group discussions, make it clear that your intention is to learn and not to judge.

• Establish confidentiality by assuring the study participants that your conversations or interviews will not be repeated to others and that when you write a report, they will not be identified by name.

Once established, good rapport with the study population can be maintained by observing certain universal rules of courtesy which include the following:

• Privacy. Investigators should be very careful about intruding in people's privacy. Private and sensitive questions such as asking to see people's latrines can cause embarrassment; so might asking about income.

However, this can be minimized if the study team ensures that the participants are well-informed about the investigators' motives and interests. If people understand why you are asking them such questions, they are more likely to cooperate with you.

• Timing. Visits to people's homes and the timing of group discussion meetings should take into account local patterns of activities. For example, investigators should avoid arriving for interviews at meal times. Make sure that you consult people before setting the dates and time for group meetings. You will need to be flexible enough to allow for possible rescheduling of activities during local holy days, market days, and other less predictable events such as, for instance, funerals.

• Feedback. Presenting the findings to the study population(s) informs those who participated in your study, and provides them with the opportunity to comment on your conclusions, to confirm your analysis, to correct any misconceptions, and perhaps also to provide additional information. Feedback helps people to increase awareness of their own hygiene practices, and to appreciate the reasons for your project's interventions. Seeing that you have taken the trouble to let your informants see the results of their cooperation, using flip-charts or summary sheets. may help them to remain interested and cooperative during the study period and beyond.

Transferring technical know-how

Training should enable the study team to learn new skills and/or improve old skills in how to use qualitative methods of investigation and analysis. It is difficult to generalize on the skills and qualifications that project staff are expected to have, as this varies greatly from one project to another. It may be that some projects have employees with social science training and/or experience, hut this is not common. Many project staff such as sanitary/water engineers, public health technicians, and health/hygiene education specialists will have considerable knowledge about environmental, microbiological, and epidemiological aspects of water supply and sanitation interventions. Those who work on the social development and educational side of the project may know more about the social, cultural, and behavioural aspects.

It is important to assess existing skills and identify areas where training and supervision should concentrate. The trainer should find out what members of the study team have already done and build on their knowledge and skills as much as possible, rather than introduce a completely new set of skills which will require more time to master.

The social scientific skills required to systematically assess hygiene practices and related issues are best learned using a combination of both theoretical and practical training methods. On the theoretical front, the trainer may give informal talks or seminars and point out key references or resource materials for the study team to read. A study resource stockpile consisting of essential reading and/or reference materials may include some of those listed in the Selected Reading list at the end of this handbook. During the initial training, encourage each member of the study team to read at least one item and to share their thoughts on it with the group. This will facilitate relatively quick assimilation of information and stimulate discussion of new ideas. On the practical front, the focus should be on activities that enhance the team's investigative and analytical skills.

The following suggestions might be included in classroom and field training sessions. The methods and tools for assessing hygiene practices described in this handbook require three main skill categories: observation, interviewing, and facilitating, moderating, or leading group discussions. Each category is discussed with reference to the particular method/ tool that requires each skill.

Observation Skills

The term observation relates not only to looking or the sense of sight, but also to the senses of touch, feel, and smell. However, having functioning senses does not automatically make a person a skilled observer. Training is required in order to know what to look for, to sharpen one's senses and also to record or document one's observations. Observers need to learn how to:

• write systematic detailed descriptions of what is observed;
• separate relevant detail from trivia without being overwhelmed by the amount of trivia.

It is often difficult to train people to be good observers because, as Patton put it, "...so many people think that they are natural observers and therefore have very little to learn. Training to become a skilled observer is a no less rigorous process than the training necessary to become a skilled statistician. People don't naturally know statistics - and people don't naturally know how to do systematic research observations. Both require training, practice and preparation." (Patton, 1990:201).

Conducting observations can demand considerable energy and concentration. The observer has to switch on her or his concentration to awaken eyes, ears, taste, touch, and smell mechanisms. You cannot expect members of the study team to conduct systematic observations spontaneously without the training and mental preparation required. Skilled observers can improve the accuracy and trustworthiness of observation data through intensive training and rigorous preparation.

However, it is important to note that even skilled and well-trained observers can bring in their own biases to the information. This may be because of the observer effect, behaving differently due to the presence of an observer. It may also be due to the inherent limitations of information derived from observation only, and not from direct experience.

For example, a mother may wash her hands thoroughly after disposing of her baby's stools during the observer's visit because she wants to give a good impression of herself. When the observer is not around, she might just wipe her hands on her dress or apron or sari, and continue with what she was doing before the baby defecated, and not worry about washing her hands. This shows that the observer effect can produce biased information. Similarly, an observer may notice that a woman collecting water from the well has covered her water pot with leaves before putting it on her head to take home. This observer may not be familiar with this practice (commonly done in several parts of Africa to prevent water from spilling while the woman carries it to her home) and may interpret it as an imprudent (unnecessarily contaminating) practice without asking the woman why she had done it.

For this reason, participant observation has in the past been seen as the best method, if not the only method to obtain indepth information on the issue under study. However, the term participant observation is an umbrella term that includes observation, interview, and discussion. The use of a combination of methods is certainly the way to minimize bias. The observer's ultimate goal is to be able to understand and explain observed phenomena as accurately as possible, but absolute accuracy may not be achievable.

Observations can vary according to focus, duration, the role adopted by the observer, and the way the observer's role is portrayed to the study population (see Box 5). Trainees should be encouraged to:

• suggest whether they would envisage assuming the role of full participant observers. partial participant observers, or onlookers from out side, and state the reasons why with reference to their particular setting;

• discuss how observations would best be conducted in the study site(s) (whether observers should let the observed know what they are doing or keep it to themselves, and why);

• suggest what the duration of observations should be, and why;

• discuss whether the focus of observations will be broad or narrow, and why.

Four Things to Consider When Planning to Conduct Observations (Adapted from Patron. 1990:217)

This will allow trainees to think about the use of observation methods analytically. Once developed, the techniques of observation can be used as integral parts of almost all the methods and tools described in Chapters 5 and 6. Direct observation is particularly an important part of conducting healthwalk and semi-structured interviews, but it is also built into the methods and tools that involve group discussion, because the interaction between participants of a group discussion is observed and noted as a matter of course in order to put in context the information gathered from that group. Structured observations carried out during home visits can provide vital information to help interpret data gathered by interviews conducted simultaneously.

Trainees may be given relatively short observations to conduct around the training place. For example, a healthwalk may be conducted in afternoon for initial training purposes. First, describe in detail participants what the healthwalk entails, and how it helps to orient the investigator to the study area. Trainees can be asked to conduct a healthwalk in the local area. This may involve a walk through the suburb surrounding the training site, or walk through a local market, or the neighbourhood of the project offices.

Trainees can be allowed up to an hour for the healthwalk, taking brief notes in a small note book. They should then return to the class/seminar room to write up more detailed notes and report to the group what they observed individually. Attention should be drawn to the variety of notes, differences in observation details, and systematic biases. For example, some people will routinely observe structures but not people; others may observe activities but not the circumstances in which activities take place, and so on.

Structured observations may be made using a checklist or tally sheet, if the items to be observed are precoded, or observations may be made systematically for the purposes of continuous monitoring and assessments of time-allocation for specific activities (see Bentley et al., 1994, for more examples).

Ask trainees to develop a checklist of things to observe in a given area. They can then make systematic observations using the checklist. They should also take notes of things not included in their checklist, or in relation to the items already included in the checklist, and later write these up. The notes might include details regarding participants, the nature of activities, and the time taken. Allow plenty of time for them to discuss their observations and the difficulties they faced.

As a result of the observation training exercises, participants' skills in observing activities, taking good notes, and awareness of observer bias should be improved. It should be emphasized that the aim is to document observed information and not to interpret or judge it on its own. Observation notes should give no indication of beliefs or attitudes of the people observed. These should be recorded separately. For example on the basis of observation alone, the study team should not conclude that women do not appreciate the advantage of a protected water source over an unprotected one. Discussions with women often reveal that water from a specific source is put to a specific use on the basis of rational decision-making processes.

Preparation and Pretesting of Observation Schedules

The preparation of a spot-check observation schedule (checklist) involves discussions of a draft checklist by the study team, translation into the local language(s), and back-translating from the local language into English in order to check the accuracy of meaning conveyed by translations. Trainees should spend some time reviewing the examples of structured (spot-check) observation schedules included in Chapters 5 and 6, and modifying them for use in their own study site(s).

Interviewing Skills

The two main skills required for successful interviewing are:

• an ability to establish rapport with the interviewee, and
• keen listening.

There are several exercises, role plays, and games that can be used to improve interviewing techniques, particularly in the areas of probing, listening, observing non-verbal cues, and recalling the content of interviews. These include:

Class Introductions

At the beginning of the training period, you can develop people's interviewing skills and establish rapport among the study team by asking them to form pairs, preferably with people they do not know well. The pair must interview each other about their life and their work sufficiently well in order to introduce each other to the group. Allow about twenty minutes for the interviews and then give each person two minutes to introduce the person interviewed.

Genealogical Exercises

Collecting a genealogy (a family history) is a good way to develop very focused interviewing skills. Again, get participants to form pairs - but different pairs from those of the preceding exercise. Get each to interview the other about family relationships, starting with the person being inter viewed. In order to build up your picture of the family (or draw the family tree), you need to ask them about parents, siblings, sibling marriages, their own marriage, children, relatives of parents, and so on. Have all trainees discuss each of these in turn. Then ask each group member to take their own genealogy, and to draw a line around the people they live with, or those who live in the same village, town, or part of town. This can also be an effective way to introduce a discussion on defining units of analysis such as family, household, homestead, community and so on.

Three-Way Interview Exercises

Ask trainees to form groups of three. Each person will in turn take on all three roles: the interviewer, the interviewee, and the observer. Choose any topic as the subject matter of the interview. The observer's role is to watch the interactions, observe possible leading questions or biases, monitor the use of prompts and probes, and note any areas overlooked by the interviewer. How well does the interviewee respond to this interviewer? After a set period of time, the three change roles, until all have had an opportunity to interview, be interviewed, and observe an interview. Ensure that there is plenty of time at the end to allow discussion of the interviews: everyone will have comments that relate to all of the roles they played, and they will learn further from listening to the feedback of others.

Saboteur

This is a game for trainees to identify ways of dealing with interruptions or break-up of communication that can happen during interviews. Trainees should again form groups of three, each taking the role of interviewer, informant/interviewee, and saboteur. The role of the saboteur, someone who engages in sabotaging the interview, can be illustrated using examples of interviewing a mother when the child(ren), the husband or a visitor is interfering in the conversation, without using violence. Examples of sabotage include interruptions. rude behaviour, such as staring, taking over the interview, and answering instead of the respondent, sitting in silence, and making noise. Each participant should have a chance to play each of the three roles and then to discuss the type of sabotage and possible ways to deal with it.

Among the ways to deal with sabotage are to:

• ignore the interruption;
• acknowledge it and deal with it immediately;
• acknowledge it and postpone dealing with it;
• ask others for help or guidance;
• involve the saboteur in the conversation;
• stop the interview.

Once the trainees have played saboteur, you may find that they will keep identifying saboteurs in non-interview settings as well, for example, those who interrupt meetings and group discussions. The same principles can be applied to deal with saboteurs effectively. This game is fun to play and can help to maintain the study team's sense of humour, particularly if faced with unpleasant or unacceptable interruptions.

All of the above exercises should help the trainer(s) identify the best interviewers in the study team. These individuals can proceed to study the example of a semi-structured interview schedule included in Chapter 6. They should review and, if necessary, modify it to suit the specific objectives of the study. Encourage interviewers to be flexible when using preset question lines. The order in which questions are asked does not matter as much as the need to maintain an uninterrupted conversation with the interviewee. The order of questions should follow logically from the interviewee's responses. For example, questions about water use can be asked first if the informant's response to the first question has mentioned water use and not latrine use.

Group Discussion Moderating/Facilitating Skills

Many of the methods/tools described in this handbook involve group discussions that require skilled facilitators/moderators. Decisions about what methods/tools to try our during the training and which ones to prepare and use in the actual study should be based upon the study objectives, the capacity of the study team, and availability of time and material resources. Avoid choosing methods and tools just because they are part of the current fad in development and/or evaluation, without assessing their appropriateness to your study.

The trainer(s) may demonstrate how to facilitate mapping by getting participants to make a map of their town showing their project office, the local clinic or hospital, the departments of health, social development, water resources, education, and so on. If the group is large, you can divide the participants into two groups and get the second group to draw a detailed map of their project site. The role of facilitator and note-taker(s) may be played by the trainer(s) or by designated trainees. Allow about an hour for this exercise and discuss the outcome, highlighting areas of difficulty.

Similarly, a historyline (see Chapter 5) can be conducted among the trainees. Encourage the more senior/experienced members of the study team to recount the history of the project, agency, or department since its inception, or as early as they can remember. Get them to mark important events and key personalities on the historyline and elaborate on their significance. About forty minutes may be allowed for this exercise including the discussion of issues raised.

Seasonal calendars (see Chapter 5) can also be conducted to find out about the activities of project staff during the different seasons. This exercise may be used to find out the best time to conduct a hygiene evaluation, from the point of view of project staff. You can then get the project staff to conduct the same exercise during a visit to a nearby area to find out what local residents' schedules look like.

Three-pile sorting and/or gender roles/tasks analysis can also be tried if you already have the materials, or can prepare them relatively easily.

Focus group discussions may be tried by getting all trainees to participate in at least one focus group discussion on a topic of interest to them all. For example, "difficulties faced in doing field work" could serve as a good discussion topic. The results of such a discussion could also help the trainer to identify areas where on-the-job training should focus. Use Part 11 of the Focus Group Manual (Dawson, Manderson, and Tallo, 1993), which is wholly dedicated to the training of project staff on the use of focus group discussions. This includes plenty of examples of how to develop question lines, the use of probing and prompting techniques, note-taking and analysis of findings.

In the development and testing of this handbook, we have used a modified version of focus group discussions in which pictures are employed to introduce and stimulate discussion on sensitive topics (see Chapter 6). A point worth emphasizing is that the purpose of focus group discussions is to encourage participants to discuss the topic among themselves and is not to get them to respond to the moderator's questions one by one. This is a very common mistake where focus group discussions are said to have taken place when in actual fact they were group interviews.

The following exercises may be useful:

• The trainees should develop some question-lines for the moderator/facilitator to use in the focus group. Stress that the question-lines should be much more general than an interview schedule - it should provide an outline only of the topics on which questions are to be asked.

• Get members of the group to take turns moderating or facilitating the group discussion on a selected topic. Choose a few controversial topics in order to make the discussion stimulating, and take an extreme role yourself if people assume a common line and reach consensus too quickly. Alternatively, give participants particular roles to play (such as an extremist, a conformist, a rebel, a modernist, and so on) and have them act out their roles during the group discussion.

Interviewing and group discussion exercises can be continued for a couple of days until the trainer(s) are satisfied with the study team's interviewing and facilitating/moderating skills.

The initial training should include guidance on the analysis, interpretation, and documentation of the information gathered. This can be introduced by getting trainees to discuss Chapter 7 of this handbook in particular and linking up the relevant issues raised in Chapters 2 and 4. However, much of the art of carrying out analysis of qualitative data is learned by doing, through on-the-job training during the conduct of investigation.

Management, review, and analysis of information

Qualitative information is analyzed as you collect it. Training should enable the study team to prepare for this time-consuming task. Try reviewing and filing your notes from the training exercises in order to get into the habit of managing data properly. Once the study begins, you will need to review the day's work at the end of each day. Write down ideas as they arise. You will need to organize your notes and put in place a good system for filing information. Advance preparation will help the study to be conducted smoothly. You do not need a computer, although computers can speed up the process of information management and review. Computers can also be hazardous as information can be lost due to power cuts, breakage, and other common mishaps that can happen in the field. In any event, encourage the study team to keep their handwritten notes in folders and notebooks. These can be stored in different ways, and retrieved easily, if a filing system has been carefully prepared.

• Keeping notes, notebooks, and files. Keeping good notes is very important, and these should not only be about things directly related to hygiene practices. Sometimes you do not know what is relevant, so you need to keep comprehensive notes of other things that you feel may be significant. It is also important from the outset of your investigation to identify new questions of relevance to your study, whenever they emerge.

• Codes and themes. Once you have written up derailed notes, mark them up to identify themes that emerge. These can be written as margin notes on the text. For example, you might use CHLDSTO to mark information about the disposal of children's stools, WATSO for water sources, DIARC for causes of diarrhoeal diseases, ACTM for mother's activities. and so on. You need to make up code words that make sense to you, so that when you see a code word, you know immediately whether the text is valuable. But you also need to keep a code book to remind you what the code words stand for. Once the study begins, and by the end of four or five weeks, you may have a lot of codes and may not remember what many of them mean, unless you keep a list of codes at the outset.

You might find a small index book, divided by alphabetical order, especially useful. Organise the code book like a dictionary, and use it only to keep a list of the codes or labels you use for your field notes, interview texts, group discussion transcriptions, and notes. You will need this to check later what certain code terms mean. Avoid complex codes that you might forget, and that others may not be able to decipher.

Developing working hypotheses

Many qualitative studies begin by formulating one or more assumptions about possible answers or solutions to the problem(s) to be addressed. In qualitative research these assumptions are called working hypotheses, suppositions based on some known facts that serve as starting points for further investigation. Box 6 provides an example of a working hypothesis used in one of the hygiene evaluation studies carried out during the development of this handbook. In this particular case, the results of the investigation confirmed the working hypothesis, but this is not a general rule.

Contents - Previous - Next