Contents - Previous - Next

4. Designing a hygiene evaluation study

Hygiene evaluation cycle
Defining the objectives of the study
Developing specific objectives
Sampling strategies
Putting in place data quality checks
Scheduling activities

Hygiene evaluation cycle

When developing this handbook, we tried to resolve what we might call the applied anthropologist's dilemma. On the one hand, we encourage project staff with little or no previous training in anthropology or related disciplines to be involved in conducting qualitative investigations. This requires practical training and conscientious supervision of project staff. On the other hand, the constraints of time and other resources which project staff face impose limitations on how satisfactorily they can engage in such systematic investigations. Very often, qualitative investigations raise issues that cannot be fully addressed in a short period of time. Can anthropological and related studies be carried out rapidly by study teams with less training than might have been desirable? The answer is yes and no at the same time: the less time available for systematic qualitative investigations, the more highly trained and knowledgeable of the study site and populations the investigators need to be.

The hygiene evaluation studies proposed by the guidelines in this handbook may be carried out rapidly in a matter of a few months, if not weeks. Rapid assessments, however focused, may leave a number of questions unanswered and point to areas where more investigation is required. For this reason, one hygiene evaluation study should not be seen as an end in itself. In a project setting, it may be helpful to visualize one hygiene evaluation study as a cycle within the larger project cycle of planning, monitoring, and measuring impact. It may be necessary to repeat a hygiene evaluation cycle (one study) periodically. The purpose of the repeat studies may be For follow-up or further investigation of issues raised by the previous one (for example, issues related to seasonal variation which cannot be addressed by a single study), or for monitoring or measuring impact. This is illustrated by linking up a series of investigations, experiential learning cycles as shown in Figure 2.

Each hygiene evaluation study (one cycle), can be seen as a four-stage learning process:

• Problem identification/defining the question(s).
• Gathering information systematically.
• Reviewing the information.
• Reflecting on the results and/or taking remedial action.

This was adapted from the principles of experiential learning, or learning by doing (Kolb, 1984). Any one of the above four points may be a starting point for the cycle, although we start with the first one. This cycle represents the different stages of a hygiene evaluation study. [The study team may also find it useful to use the same principles in their daily routine sessions of information gathering, review, interpretation of results, identification of questions for further investigation and information gathering, etc. throughout the study period.]

If the purpose of your study is to measure the impact of project intervention, it may be necessary to add a quantitative component to your study in addition to the qualitative investigation. In that case, you may consider the use of a questionnaire that has been formulated to reflect the findings of your qualitative investigation.

At the end of your study, you will have defined a number of issues to be tackled in the next rapid assessment cycle, that will help complete the picture obtained by the present study. If you are able to follow the cycles of investigation/evaluation, implementation and further investigation in your project work, you will have resolved the problems of incomplete results and limited resources, benefitting from a core study team trained in a piecemeal approach to systematic qualitative investigations.

Defining the objectives of the study

The objectives of your study will first depend on its purpose. If, for example, you want to assess the effectiveness of an intervention you have carried out to change particular hygiene practices, then your objectives will include the measurement of change in these specific practices. If, however, you intend to obtain baseline information on existing hygiene practices prior to developing a successful hygiene education intervention, you may want to study a greater range of hygiene practices in order to find out which are most in need of intervention. The overall objectives should be decided in consultation with the expected users of the information gathered in the study - for example, the project managers - and as much as possible with representatives of the study population(s) as well. Study aims and general objectives are closely linked to the study's intended outputs. Box 7 provides an example in which these were defined during the preplanning and initial training phase of a hygiene evaluation study conducted in Kerala. India.

When defining study objectives, you need to consider which cluster or clusters of hygiene practices to investigate (see Table 2). The number of clusters to consider will be influenced by the purpose of your study and available time and other resources. If time is limited, it may be best to prioritize and select one or two clusters rather than risk the quality of your study by trying to address too many clusters in a short period of time. For example, in a project where improved water supplies and sanitation are being introduced, it may be preferable to focus on excrete disposal (Cluster A) and water uses (Cluster B),including water sources (Cluster B) if time allows. Such choices need to take into account local conditions, particularly by considering where changes could be most influential in improving the health of the population. It is important to look to the future as well as the present, and link the results to follow-up action that is indicated for the project itself, where changes may need to be made. Plans for future investigations should also be considered (see "Hygiene Evaluation Cycle" in Chapter 4).

Table 2. A guide to the five clusters of hygiene practices

Developing specific objectives

A choice must be made about which specific hygiene practices will be investigated within each cluster (see Table 2). Within excrete disposal, for example, latrine maintenance and cleanliness, disposal of children's faeces, and hand-washing after defecation could be selected as practices most influential in the prevention of disease. Such choices need to be made with some knowledge and understanding of the local hygiene situation (see Box 8 for an example of specific objectives developed for one of the overall objectives described in Box 7). Note that the selection of specific hygiene practices made while preplanning and designing the study will be provisional, as you may need to modify or change your list once investigation starts and you discover important practices which had not been identified previously.

Your choice of hygiene practices should be influenced by which water/ sanitation-related infections prevail in your study site and which specific practices may be associated with the transmission of these infections. For example, the World Health Organization Control of Diarrhoeal Diseases Programme suggests that it is important to focus on three sets of hygiene practices, rather than include all possible transmission routes. These are:

• safe disposal of human excrete, particularly the faeces of young children and babies, and of people with diarrhea;

• hand-washing, after defecation, after handling babies faeces, before feeding and eating, and before preparing food;

• protecting drinking water from faecal contamination, in the home and at the source (WHO, 1993a:8).

It may be possible to investigate the particular practice of interest, for example, water collection and handling practices at the source, by observation. An observer may go to the source and observe people's activities without attracting too much attention. Other hygiene practices and behaviours are harder to observe, for example, adult defecation, as this is a private matter. In such cases, we suggest that indicators or signs of behaviour be observed. A number of indicators that are both effective in indicating the occurrence of a particular practice, and relatively easy to assess have been used. These include:

• Cleanliness, safety, and soundness of latrine structure. Indicators of use of latrine. People are unlikely to use one that smells, has a shaky and/ or dirty floor, a leaky roof, or is surrounded by bushes where snakes like to nest.

• location of latrine in relation to living quarters. Indicator of use of latrine. One that is situated too close to or too far from the living quarters, or on the wrong side of the courtyard, or on publicly owned land, may not be used much or may not be maintained well for reasons that will be worth investigating.

• Means of disposal of children's faeces. Absence of children's faeces in and around the home and/or presence of potties or clean latrines in households indicates the isolation of faecal contamination in and around the home. This can be investigated relatively easily because attention is drawn to children's rather than adults' hygiene practices, to minimize embarrassment.

• Turbidity and smell of water at the source and in the home. Indicator of contamination of water source and lack of protection. Water that looks muddy or unclear at the source and water that smells of rotting leaves or other organic matter may provide clues of contamination. However, it is necessary to observe how it is used (for example, whether it is filtered, or clarified by adding ash or other materials, or boiled or used for non-drinking purposes) and to ask questions about what is observed.

• Presence of soap/hand-washing facilities near latrine. Indicator of handwashing after defecation.

Sampling strategies

Sampling is as Important to information gathering as It IS to the analysis and interpretation of findings. As Miles and Huberman (1994:27) put it, "as much as you might want to, you cannot study everyone everywhere doing everything. Your choices (whom to look at or talk with, where, when, about what, and why) all place limits on the conclusions you can draw. and how confident you and others feel about them." Sampling related questions that are frequently asked when designing a hygiene evaluation study include:

• How many people should I include in my study?
• What criteria should I use for recruiting participants to the study?
• How many interviews, observations, and group discussions will I need to conduct?

The short answers to these are:

• As many as can provide adequate answers to your questions.
• Criteria which reflect the objectives of your study.
• As many as can provide the information you are looking for, within the limit of resources (human, material, and time) available to you.

Sampling is perhaps the clearest dividing line between quantitative and qualitative investigations. In particular, two sampling-related differences are noteworthy:

• Qualitative investigations employ small sample sizes where a relatively small number of people set in their context are studied indepth. This is unlike quantitative investigations where large numbers of context free cases are studied in search of statistical significance.

• Qualitative samples tend to be, in the main, purposeful and not random. This is partly because the initial definition of the whole is more specific, and partly because social processes have a logic and coherence that random sampling would miss out completely.

In quantitative surveys, standardized scales are used so that individuals and groups can be described as showing more or less of some characteristic, for example, knowledge. Everyone is rated on a limited set of predetermined dimensions. Statistical analyses of these dimensions emphasize central tendencies - averages and deviations from those averages. By comparison, qualitative investigations pay particular attention to uniqueness, be it of the individual, the household, or any other unit of analysis. For this reason, the scales used are not standardized. Instead, they are adapted to take individual variations into account, while being sensitive to similarities among people and generalizations about them.

Sampling strategies may appear complicated at first sight. However, this need not be the case if you allow some time for careful planning and detailed consideration of what it is you would like to say something about in your study. As Feuerstein put it simply (1986:69), "sampling means looking closely at part of something in order to learn more about the whole thing." If, for example, you want to find out what the food in a cooking pot tastes like, you can take a spoonful and taste it. You do not have to eat all the food. This type of sampling is fine if there is only one pot of food to taste. However, when discussing sampling from our study population, we need to think in terms of several pots of food which make up a meal. Each pot has to be sampled to find out what the various parts of the meal are. You can then say something about the food in each pot, as well as the meal as a whole. There are different types of sampling that can be applied when studying a population or a community that is made up of diverse component parts, such as ethnic groups, age groups, etc. which may further be differentiated by gender, socioeconomic status, etc. It is important to distinguish between purposeful and random sampling strategies (see Box 9). The two types of sampling serve different purposes. Often, combined or mixed sampling strategies are employed within the same study in order to answer different questions. For example, purposeful samples are often not prespecified at the beginning of the study. They can evolve during the study. Initial choice of informants/ study participants may lead the investigator to similar and different informants; observing one cluster of hygiene practices may invite comparison with another cluster; and understanding one key relationship in one setting may reveal issues to be addressed in other settings.

More than one sampling strategy can be applied in two different ways: working from the outside into the core of a setting, or working from the core into the wider context of the issue at hand. For example, in studying hygiene practices, you may begin with the study population (use census data, walk around the settlement areas) and then enter the selected villages or sections of towns and the households, staying several days to get a sense of the frequency and occurrence of different events. From there, the focus would be on specific events, times and locations. Alternatively, you may observe a specific practice in one locality and work towards investigating the wider context by looking at neighbouring and more distant localities (see Miles and Huberman, 1994 and Patton, 1986 for more examples and detailed discussions of qualitative sampling issues).

In summary, there are two important steps you need to follow when deciding how many people to include in your study and what criteria you should use for including them:

Step 1: Define the population boundaries of your study site. You may have existing boundaries to distinguish areas where your project activities are currently going on from those that have not had any intervention yet, or those where your project work has already finished. Alternatively, you may find it more useful to define your study population according to political/administrative boundaries such as region, district, division, location or sub-location; according to income levels such as high, middle, and low-income.

Step 2: Find out how many different parts or groups make up the whole, and determine their relevance to the questions your study is aiming to address. You can then take samples or representations of the whole to include in your study.

The example in Box 10 illustrates how samples were recruited according to the purpose of the study, the type of questions addressed, and the indicators used in a hygiene evaluation study. In this case, the working hypothesis was that faecal contamination in and around the home and the hygiene practices associated with them are more prevalent and therefore easier to observe in homesteads with young children than in those without. Participants to group discussions were invited and grouped purposefully.

Contents - Previous - Next