Previous - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
Ethnographic studies of the effects of food availability and infant feeding practices
Gretel H. Pelto
Department of Nutritional Sciences, University of Connecticut,
Storrs, Connecticut, USA
INTRODUCTION
In recent years systematic research in social epidemiology has confirmed what public health professionals working in the field have often observed, namely, that the causes of infant malnutrition are complex and vary from one community or region to another. Understanding the main factors that affect infant nutrition and health in a particular environment may help to identify effective "intervention points" for programme development and lead to more efficient utilization of resources for the prevention of the "vicious cycle of undernutrition."
In public health research the usual method for collecting data is the large-scale survey, which typically includes questions on nutrition and health status indicators, as well as some social and economic data. This paper outlines a series of information collection procedures that can be used (a) to improve the quality and utility of a large survey by identifying specific, locally significant variables and suggesting culturally appropriate ways of asking questions in the survey, or (b) to replace survey methods when resources are so limited that only the most minimal information collection can be undertaken prior to programme development. Given the obvious advantages of a survey to elaborate and confirm the initial generalizations derived from the procedures outlined below, the paper is written from the perspective of the first use rather than the second.
The paper outlines a series of questions to be asked concerning factors that may affect infant feeding, nutrition and health. It also directs attention to background factors and characteristics of the community that may be important for programme development. The information can be collected through several techniques: observation in the community, informal conversations, and structured "key informant" interviews. The extent and depth of the interviews, the time expended, and the number of people contacted will depend on resources available and the plans for a subsequent, more formal survey. We have used the term "ethnographic work" to refer to this type of information collection because the techniques have been derived from techniques characteristic of anthropological field research.
In deciding what information to collect, one can be guided by a number of propositions focused on some of the primary determinants of food intake and nutritional status:
1. The nutritional status of an individual (infant, child, or adult) is strongly affected by conditions and characteristics of the household in which that individual lives. Therefore, it is necessary to obtain information about the household.
2. The material resources of a household (land, income, and so on) should be regarded as a primary influence on nutrition through their effect on food availability.
3. Cultural beliefs and values affect decisions about food acquisition, preparation, and consumption. Therefore, it is necessary to obtain information about the cultural aspects of food use.
4. The synergistic relationship between nutrition and infection is of such magnitude that, at least in areas of serious, endemic infectious disease, it is necessary to obtain data on morbidity as well as on food intake.
5. Infant-feeding practices are best seen as an aspect of child care, and it is therefore important to understand the context of child care (including maternal activities and time allocation) in order to comprehend the logic and patterning of these practices.
From these assumptions, a general model of the determinants of nutritional status of infants and young children can be derived. This modes postulates that nutritional status is the outcome of interactions among a series of variables that can be classified under the following general headings:
GUIDELINES FOR ETHNOGRAPHIC WORK
The field-work activities should be concentrated in two areas: general community description and data on the sectors of information pertaining to the household, following the outline above. With respect to the latter, it is assumed that one purpose of the work is to improve the design of a survey by identifying key questions, finding what might be threats to the validity of subjects' responses, and providing guidelines to reduce a household interview schedule to a manageable length.
General Community Description
Although broad description of the communities in which the information collection takes place is not a primary aim of the ethnographic work, there is a minimum data base that should be accumulated and elaborated to the extent that time and personnel resources allow. Census data and other available data should be examined prior to beginning the descriptive work. It should be stressed that the sectors of data for community description that follow are to be collected at the level of the community, not at the level of the household. The guidelines are developed for a rural community. The general community description should include the following:
Sectors of Information about the Household
The sections that follow are divided into the five main areas or sectors of information that are the determinants (or independent variables) to be linked with infant and young child nutritional status. Each section outlines a series of questions and sets out specific research tasks. The questions are intended to serve as guidelines for the ethnographic observation and interviewing. It will be impossible to obtain even preliminary answers to all of them in a short time; nevertheless, to the extent possible, some data (however minimal) should be accumulated to answer the questions. It is important to cover all sectors, and so detail within any given sector may have to be sacrificed in the interests of broad coverage. On the other hand, another main purpose of this phase of work is to discover new information and generate new hypotheses. Thus it is important to regard the following sections as suggestions to assist the research process rather than as a rigid framework.
INFANT FEEDING AND CHILD-CARE PRACTICES
This section concerns the heart of the research-the factors that are immediately proximate to infant and young child nutritional status. It is organized in two parts: a series of questions concerning relevant behaviours and beliefs, and a research plan for obtaining this information, the format of which can be applied to the other sectors of information gathering.
Questions on Behaviours and Beliefs
Birth and the Early Postnatal Period
To understand feeding practices in the immediate postnatal period, it is important to have some information on pregnancy and childbirth, including data on the social support and dietary practices affecting women during this potentially stressful time.
1. Where does childbirth occur-in the home of the couple or in the mother's girlhood home? Are there differences in birth location of second and subsequent children compared to firstborn? What is the effect of distance from mother's natal home on birthplace? If childbirth takes place in the mother's natal home, how long does she remain there before returning to her marital residence?
2. If childbirth takes place in the marital home or village, do any of the mother's kin come to stay with her? Is there a difference in first births compared to subsequent births with respect to this?
3. Who attends the mother at birth? To what extent do women rely on traditional birth attendants?
4. How much do women use prenatal medical services? How often do deliveries take place in a clinic rather than at home?
5. Is the baby put to the breast immediately after birth or after some period of time? Does this differ with the location of delivery (e.g., home versus clinic)?
6. Is colostrum distinquished from milk linguistically? How is colostrum regarded (what are its properties)?
7. Is a newborn put to the breast of a woman who is already in full lactation rather than, or in addition to, the breast of the mother? If so, what is given as the rationale for this practice?
8. Are any special foods/drinks given to the new mother to promote milk production? Are there any foods that a newly delivered mother should avoid? Why?
9. What are women's expectations about problems associated with the early postnatal period? Do they expect to have sore nipples or breast engorgement? Are such conditions regarded as normal or abnormal? What are the remedies?
10. In the early postnatal period, are babies put to both breasts at a feeding or only one? What is the behavioural pattern of breast-feeding?
11. Are infants given anything else in addition to (or in place of) breast milk in the early postnatal period? What? How is it prepared? How is it given?
Perception of Developmental Stages
In order to develop questions about changes in infant feeding and child-care practices as the child matures, it is important to understand how people perceive developmental stages. If there are clear, culturally recognized markers of development (e.g., "when a child sits," "when a child talks," etc.), these can be used to inquire about feeding practices.
1. What behavioural and/or physical changes in infants are recognized as important?
2. Are any of these developmental stages given names, such as "lap child" or "yard child"7 In other words, are there cultural equivalents to the Western distinction between infant and toddler?
Perceptions of Breast Milk, Other Milk, and Water
1. What do people regard as the qualities of breast milk that distinquish it from the milk of other animals? Is commercial infant formula recognized as modified cow's milk or as a separate entity?
2. Can breast milk "spoil," or become bad for a child? What causes the milk to be bad? How does one know that it is bad?
3. Is there a concept of milk insufficiency-i.e., do women recognize the possibility that milk volume is low? Are concerns about volume linked to the child's developmental stage?
4. Are babies given milk (other than human milk)? At what developmental stage? Is this a routine practice, initiated as part of weaning, or does it occur only under special conditions (e.g., mother's illness or perception that the baby is not satisfied, etc.)?
5. How frequently are infants given water or other liquids? Does this change with developmental stage? How is the water prepared? How is it given?
Infant Food and Feeding Practices
Since infants and young children are rapidly changing organisms, the following questions must be asked with respect to particular developmental stages. For recognized (post-neonatal) stages (or some clearly defined markers, such as "when a child crawls"), the following types of information should be obtained:
1. What food(s) can or should be given to a child of this age? What foods should not be given? Why?
2. How are these foods prepared? How are they given to the child?
3. How do you know that a child (of this developmental stage) is ready for (food)?
4. Is a child (of this developmental stage) ever "force fed"? Under what circumstances? What kind of food is forced?
5. How does one know when a child (of this age) is hungry? Thirsty? What characterizes a child who has eaten enough?
6. How many times per day is a child (of this age) fed? Is this pattern maintained when the mother is away from home for a day or part of a day?
7. If a mother takes a baby (of this age) with her, away from home, does she carry food for it? Buy food? Not give additional food?
8. What changes in feeding practices occur when a baby (of this age) is sick? What foods are withheld? Are any special foods given? Do different types of illnesses call for different dietary changes?
9. What types of changes in feeding practices occur when the mother of a baby (of this age) is sick? Do different types of illnesses (of the mother) lead to different types of dietary change for the baby?
Caretakers
1. For each developmental stage, where does the baby stay during the day? At night?
2. At what age does an infant accompany his/her mother when she leaves the house?
3. What kinds of differences in child-care practices occur in different types of households (i.e., of different household composition)?
4. At what age is an infant left in the care of someone other than his/her mother?
5. At what age is an infant entrusted to a child caretaker, as contrasted with an adult caretaker other than the mother?
6. What is the role of the father in infant care?
7. What are the normative expectations about infant care? Are there generational differences in these expectations?
Information Collection Plan for Infant Feeding
The following sets of activities and operations are intended as general guidelines to assist in the development of answers to the preceding sets of questions. The same techniques can be used to obtain a data base for the other sectors of information-gathering.
Observation: Note carefully and make a case record of all of your observations of infant feeding. Keep a file of cases in which you include as much information as possible: location, approximate age of mother, approximate age of baby, type of food, how given, if breast-feeding, thorough description of mother's and baby's behaviour, mother's position, baby's position, duration, and so on.
Key-informant structured interviewing: Identify several women (of different ages) who are articulate and knowledgeable. After establishing good rapport and making clear your interest in infant feeding, more formal interviewing can begin.
Open-ended interviewing: Take every opportunity (e.g. when a baby is being fed) to ask a primary set of questions, derived from the list above.
Small community sample: Once you have a general knowledge about the main issues and the types of answers you can expect, select a small sample of eight to ten women and interview them with a structured interview schedule that includes questions on infant feeding practices and beliefs.
FIG. 1. Social Factors Affecting Nutritional Status
HOUSEHOLD DIET AND FOOD PATTERNS
While household economic conditions provide the context for good or poor nutrition, the relationship between nutrition and social factors involves a series of linkages (fig. 1). Although these linkages can be regarded as a causal chain, each component is itself influenced by multiple factors. For example, nutritional status is only partially a function of food intake, because factors such as disease, physiological state, and energy expenditure are also involved.
At present there is relatively little systematic knowledge about the relationship of intra-household food distribution and total family food availability to infant-feeding practices. The following sets of questions pertaining to various components of food availability and family food practices are intended to help elucidate the relationships among these factors. Specifically, the purpose of the questions about land and crops, food storage, and food acquisition is to discover the best way to assess household food availability.
Food Availability
Land and Crops
1. What are the forms of land ownership? Is there any renting, leasing, or sharecropping?
2. What determines how much land an individual or household farms?
3. What types of crops are grown?
4. What are the agricultural activities through the year? What appear to be the factors that influence interhousehold and intercommunity differences in agricultural activities?
5. For each crop the following questions should be considered:
Food Storage
1. Following harvesting, how are crops processed prior to storage?
2. For each crop, what are the forms of storage? Where are crops stored? Where are purchased foods stored?
3. Are seeds saved for the next planting season? Are these set apart in a different storage area?
4. Do households save grain for times of food scarcity?
5. Can food stores be readily assessed? If not, are there any indirect measures of food stores?
Food Acquisition
1. What is the full inventory of available foods in the community? For each food: is it entirely home-produced, partly home-produced and partly purchased, or usually purchased?
2. Are there any wiId, gathered foods? What is the seasonal pattern of gathered foods? Who uses them?
3. How are home-produced animal products (eggs, milk) utilized in the household?
4. How often and under what circumstances do people kill animals for meat? Is the meat consumed at home or sold?
5. Does the sale of animal products take preference over home use or vice versa?
6. In a household, who purchases food in the market? Are some food items purchased by men, others by women?
7. Will people report accurately how much money they have spent in the market for food? Are there any tendencies to systematic distortion of reports of market purchases (e.g., particular types of food or amounts)?
Synthesizing the information: Create a food inventory. For each food include usual and alternative modes of acquisition, form of storage, seasonal availability, price, and seasonal variation in price.
Food Utilization
Food Preparation
1. Who decides what should be prepared on an ordinary day? On special days?
2. What is the inventory of food preparation tasks? Who is responsible for each of these tasks?
3. At what age do children begin to have responsibilities for food preparation? What are the responsibilities of girts? Are boys given any food preparation tasks or responsibilities for food preparation?
4. If there is more than one adult woman in a household, do they trade off food preparation responsibilities?
5. What is the full inventory of types of food preparation equipment to be found in the community? What is the cost of each item? What is the minimal set that all households have? What are the items that are optional or additional?
6. What types of fuels are used in cooking? What are the relative costs in time and money of procuring these different types?
7. Where are food preparation areas located? Are there any differences among households in the location?
8. Is there a mechanical grain mill in the community? Where is it located, and what is the cost of using it? Is there intra-community variation in the use of mechanical milling? How much preparation takes place before milling? Are grain preparation techniques the same prior to mechanical as contrasted with hand-milling?
Food Consumption
The questions below are intended to provide the basis for a description of food consumption patterns and to assist in the identification of questions for a household survey. In view of the second purpose, the field-worker should be particularly sensitive to any information that may help to improve the collection of dietary intake data.
1. Who usually eats together, and where is food consumed? If men and women eat separately, is there evidence to suggest that children eat differently when they eat with their mother than when they eat with their father?
2. What are the usual times of day for eating? Are there names for particular meals? What are the usual foods for these meals?
3. Is there variation between households in the number of times per day that meals are eaten?
4. Is there variation across seasons of the year in the number of meals?
5. Do people make a distinction between meals and snacks? What foods are eaten as snacks?
6. Are there any changes in the timing and content of meals across the week? Does the effect of day of the week differ in agricultural versus non-agricultural households?
7. Does market day affect home food consumption? What kinds of snacks are consumed in the market and not brought home?
8. How are specific foods served? Do individuals have their own plates or do they eat only from a common pot? How much variation is there, from one household to another, in the number and types of serving vessels and utensils?
Synthesizing the information: Write a description of food consumption patterns, based on the preceding questions, and provide a set of recommendations for undertaking the dietary survey.
Food Ideology
The questions in this section pertain to the cognitive cultural aspects of food consumption, except for beliefs related to infant feeding, described earlier. This set of questions is for the initial fieldwork, but the development of more in-depth understanding will require a longer time in the community.
1. What verbs are used to describe eating and drinking? What are the distinctions among these terms?
2. What adjectives are used to describe foods? What are the main dimensions used to classify food?
3. Are there any foods that are regarded as particularly appropriate for various categories of persons (e.g., children, males, old people)? Excluding infants, are there foods that are regarded as especially healthy for anyone to eat? As healthy for particular categories of persons? As healthy under certain circumstances (e.g., balance of "humoral qualities")?
4. Are there any foods that are proscribed during pregnancy? What problems are thought to be associated with their consumption?
5. Are there any foods that are prescribed during pregnancy? What benefits are they thought to confer?
6. What are the beliefs about the relationship of food consumption to illness? Can eating certain foods cause one to become sick, or to recover from illness? Specifically, for particular illness states, what foods are proscribed and/or prescribed?
7. Apart from illness or pregnancy, are there any food restrictions or food taboos?
8. How is food used in relation to ritual and religious activity? How does religion affect food consumption?
9. What foods are the most highly valued? For each food, what is the basis of its value? Are there any foods that are regarded as indispensable for particular types of social occasions (what foods and what occasions)?
10. Are there any foods that are regarded as "starvation food," to be consumed only in times of scarcity? Are there any other lowprestige foods? Will people report accurately about their consumption?
HOUSEHOLD SOCIAL CHARACTERISTICS
Household Composition and Structure
From the growing body of research in nutritional epidemiology, it is becoming increasingly clear that family composition is a very important factor affecting the nutritional status of children. For example, the ratio of children to adults, the birth order of a child and birth interval, and the temporary or permanent absence of family members (especially male household head) have been repeatedly associated with differential nutritional status of infants and children in many different cultures. Thus, it is necessary to design questions to obtain information on household composition and structure.
In situations of multiple ethnic groups, the definition of the household has to be determined for each group. In extended or polygynous households, the relationship of living quarters to a shared food supply is a general issue on which information should be accumulated. The understanding of food exchange between households also rests on the definition of household, a further reason to develop a good cultural description of the household.
1. What are the terms to use to ask about household composition? (In the survey it may be best to reference household members in relation to the index child-father, mother, half-sibling, father's mother, etc. Therefore, it is important to elicit a full set of kinship terms in the local language. )
2. With respect to kinship and non-kin relationships of individuals within the household, are there any sensitive or problematic relationships that will be difficult to elicit?
3. If there are no birth certificates, baptismal records, or other written records, what is the best way to obtain people's age? (With adults it may be difficult to obtain age in exact years, in which case a general estimate will have to suffice. To facilitate age estimation, it may be useful to have a calendar of major community and national events for the past 70 years.)
4. What is the best way to get exact age of children under two? (An adequate assessment of a child's nutritional status depends on knowledge about the child's age, at least to the nearest month. This question therefore requires special effort.)
5. What are migration patterns into and out of the communities? How often is household composition affected by short-term migration? Who leaves? Where do they go? For how long? How frequently do individuals who migrate out for work return home?
6. How fluid or stable is household composition? How often is household size augmented by long-term visitors?
7. Are there any household members or household residents who are likely to be omitted from an elicited enumeration of household membership? (E.g., are mentally or physically handicapped individuals, informally adopted children, or other categories of persons likely not to be included in response to a question on household membership? )
8. Beyond the members of the immediate household, what is the variation in the proximity of kin? What delimits the family? How difficult or time-consuming would it be to collect information on the larger kinship unit?
Maternal Activities
Through observation and informal interviewing, a list of the types of activities engaged in by women should be constructed. Once such a list is developed, the following questions can be examined:
1. What is the appropriate time unit of recall for activities? Can women report accurately for a week?
2. What is the average time and the approximate range of time that it takes women to complete each of the major tasks?
3. Do women perceive that the time it takes to complete a task changes during pregnancy or lactation?
4. What do women regard as the impact of seasons on types of tasks and amount of time it takes to complete particular tasks?
5. Do women conceptualize tasks in a hierarchy from essential to optional? Are there any tasks that are seen as optional during pregnancy and/or lactation? What tasks do women omit when they are not well?
6. What activities do women regard as recreational? What tasks are regarded as particularly onerous?
7. What tasks can women delegate to children? At what age? What tasks are shared by adult women in a household? What is the best way to ask about the allocation of tasks to others?
Socio-economic Status
To recognize that malnutrition in developing countries is first and foremost a result of poverty is to say that household resources establish the conditions for individual nutritional status. The concept of household resources encompasses not only resources for food production and purchase, but also educational resources and patterns of resource allocation. Resource allocation, in turn, is influenced by a series of factors, including degree of utilization of, and desire for, products that are associated with urban life-styles. Thus, for ethnographic investigation, socioeconomic status can be broadly defined to include direct economic resources, social prestige (and social obligation), education, resource allocation, and degree of exposure to urban life-styles.
Material Style of Life
In many societies it is very difficult to obtain data on cash income, and it is often not a good measure of economic resources even when it can be accurately determined. On the other hand, the material possessions of a household are often easier to elicit; they reflect past monetary expenditures and household resource allocation. In most communities in the modern world there are differences in household possessions and physical characteristics of the house from one household to another. A composite index of possessions and characteristics can therefore serve as a culturally appropriate measure of socio-economic status.
At the same time, it should be recognized that some possessions have high social value or are social necessities, independent of their monetary value. For all items involved in the questions below, efforts should be made to determine the monetary value, the social value, and the prestige value. The purpose of the following questions is to direct attention to the identification of items for a material-style-of-life index:
1. What kinds of possessions are found in the community?
2. How are they acquired (home-manufactured, purchased, usually given as gifts, etc.)?
3. For each item, what is its monetary value, social value, and prestige value?
4. What possessions are used for agricultural production? Are there any items that are used exclusively or primarily for cash-generating activities as contrasted with production for home consumption?
5. For particular categories of possessions (e.g., floor mats, enamel trays) are there differences in monetary or prestige value of particular styles or materials?
6. For what items is quantity rather than quality the main issue?
7. What are the differences in housing characteristics that reflect economic, as contrasted with ethnic, distinctions?
8. What aspects of house construction or housing characteristics reflect differences in household age rather than strictly economic differences?
Synthesizing the information: From the data collected, construct a material-style-of-life index that can serve as a main measure of economic status.
Cash Income
The feasibility and meaning of a measure of cash income should be assessed during the initial ethnographic phase. The following questions may serve as a guide for this effort:
1. Will people give information to an interviewer about their cash income?
2. Do the answers reflect reality? Are they valid?
3. Are there any good surrogates for cash income, such as the number of times agricultural products have been sold in the market or the amount of land devoted to a cash crop?
4. Apart from agricultural production, what are the other sources for cash income?
5. If you know someone's non-agricultural occupation, can you estimate the income?
Education
1. What is the variation in level of education within the community? Is there a sufficient range to make it worthwhile to ask about level of schooling?
2. Will people tell an interviewer accurately the amount of formal education they have had?
3. Is there any stigma about not being able to read or write, or is it easy to obtain information on literacy of individuals?
4. Is literacy a better measure than years of schooling?
Exposure to the Outside World
In societies that are undergoing rapid social change, differences in level of exposure to life-styles and information about events outside the local community are often significant factors affecting food selection, utilization of health services, and decisions about how to allocate time and money resources. Although exposure and information are not primary variables in this study, to the extent that indicators of cosmopolitanism can be easily identified, it may be useful to include this in a survey because a number of studies have shown that modernization affects nutrition. Throughout the ethnographic work, investigators should be on the lookout for ways to measure this variable.
Prestige
In many societies an individual's prestige is closely associated with his wealth. However, it is also important to recognize that in many cultures it is possible to enjoy one without the other. Sorting out the contrast often requires careful work, which would be beyond the scope of this study unless one were interested in testing the following hypothesis: "Families with higher prestige, independent of wealth, will have children with better nutritional status." One could also hypothesize an inverse relationship in situations where traditional feeding practices are not optimal and adherence to cultural traditions is a mark of prestige. To test this, one would have to rate families on prestige, a task that might require the assistance of key informants. If, in the course of the ethnographic work, this concept seems to be important in the community and seems to affect food availability and/or infant feeding practices, then this topic deserves further work.
Ethnic and Religious Identity
Cultural and religious norms and expectations influence child-rearing practices, including infant feeding, in all societies. For example, the Koranic injunction to breast feed for two years and the widespread tradition in West Africa prohibiting intercourse while a woman is nursing a child can be expected to affect infant-feeding practices, even when the cultural rules are not strictly adhered to. In multiethnic and multi-religious communities, cultural differences may be significant in determining nutritional status of infants through influences on infant feeding, child care, illness management, and related factors. At the same time, it is important to note that beliefs and expectations are not uniform within a cultural group. Intracultural diversity in beliefs and practices are to be found even in very small, isolated communities. Thus, it is necessary to identify the range of variation within a culture as well as to describe the main components of the belief system.
In societies undergoing rapid change from traditional to modern (urban-influenced) life-styles, ethnic identity can be thought of as a continuum ranging from a strong commitment to the traditions of one's ethnic group to marginal identification as a member of the group. The following questions should be considered, always keeping the matter of intra-cultural diversity in mind:
1. Do people usually give an honest response to the question of their ethnic identity?
2. How often does marriage across ethnic groups occur? When parents come from two different groups, how do people respond to a question about their ethnic affiliation? Are they likely to report their association with both or only one?
3. Does length of residence in the community affect responses to questions about ethnic identity?
4. Are there any characteristics that are locally regarded as clear markers of particular ethnic groups?
5. What is the appropriate way to ask about religious affiliation?
6. What are the markers of religiousness (degree of commitment to participation in one's religion)?
7. Are there any specific religious precepts that directly or indirectly affect infant-feeding practices?
ECOLOGICAL CHARACTERISTICS
In a community one can expect to find differences from one household to another in the magnitude of a child's exposure to pathogenic agents, differences that are the result of a number of factors, including household access to water, defecation and bathing practices, and number of animals and their care. The questions in this section are intended to help identify and define these ecological variables related to health.
Water
1. What are the types of water resources in the community at different seasons of the year?
2. What types of containers are used to store and carry water?
3. Within the community, what is the variation in distance to and access to water?
4. Are there any variations in treatment of water in the home (e.g., boiling, filtering)?
5. What factors seem to affect water availability at the household, as contrasted with the community, level?
6. How frequently, and in what manner, do people bathe? How is this affected by season of the year?
Synthesizing the information: Write a general description of water utilization that includes patterns of use (e.g., conditions under which water is reused), intra-community range in access, range in effect of seasonal contrasts, and so on.
Animals
Although animals are included here under the heading of ecological characteristics, they are also related to socioeconomic status. It should not be assumed that possession of many animals is, ipso facto, a measure of wealth, because in many communities herders' house holds are among the poorest. Moreover, animal ownership may influence the level of exposure to disease. The following questions pertain to economic and health issues related to animals:
1. From one household to another, what is the range in number and types of animals owned?
2. What is the monetary value of different types of animals?
3. What types of animal sheltering are practiced in the community?
4. Is there a range of variation in proximity of animals to human eating and sleeping quarters of such a magnitude that this should be measured in a survey?
5. Does the placement of grain storage facilities relate to animal-keeping practices?
6. Who takes care of animals?
7. What are the conditions or opportunities for exposure to animal faeces?
8. Do people perceive animals or animal waste as potential sources of contamination?
Human Waste Disposal
1. Do any households have latrines? If so, when were they built?
2. What are the usual places for defecation?
3. In what manner is it most feasible to ask about (or observe) the household arrangements for defecation?
4. Are there any problems with asking people about toilet training practices?
5. How is infant defecation and urination handled? Are there any changes in practices when a child has diarrhoea?
6. Is hand-washing after defecation practiced? After cleaning a baby?
INFANT AND FAMILY MORBIDITY
Another major factor influencing nutritional status is the number and types of illnesses the infant experiences. Thus, it is important to collect data on illness episodes and treatment regimes, since the latter may affect the duration and severity of the illness. In order to collect accurate information on illness and its treatment, it is necessary to understand how people perceive illness. Ideally, one would like to have data on the occurrence of illness in the household, not just in the index child; however, this may not be feasible. Minimally, it will be important to understand how illness in the mother affects the index child. The following questions address various components of data gathering on illness.
1. In the local language(s), what terms are used to label the main illnesses of infants, children, and adults and illnesses during pregnancy? What are the attributes or characteristics of these illnesses?
2. For each illness what are the causes? What kinds of home remedies are utilized? What therapeutic techniques do traditional healers use to deal with the problem? Under what circumstances will people seek biomedical care for the illness, and what do they expect the health care workers to do?
3. What kinds of practices are used to prevent illness in infants? During pregnancy? In childhood and adolescence? In adulthood and old age?
4. What is the history of utilization of biomedical care? Where do people obtain care? What are the types of attitudes expressed about traditional healers and biomedical practitioners?
5. Are there any beliefs concerning the causes of illness that may affect data collection in the communities?
6. From previous studies or other experience in the area, what do people perceive about infant-weighing and anthropometry? Have people been exposed to them?
7. Are there any beliefs about illness that will affect a medical prophylaxis and health care programme, including its acceptability, management, and compliance?
UTILIZING THE ETHNOGRAPHIC MATERIALS
Key-informant interviewing and observation, following the guidelines and topics listed above, will produce a wealth of data. These data can be used in a variety of ways. The information provides a solid basis for structured interviews in a large-scale survey. With greater knowledge about locally pertinent details it is often possible to reduce the length of a structured interview while maintaining a rich data collection base. The appropriate language of interviewing also can be derived from these qualitative materials.
Key-informant interviews, together with some well-selected family interviews, can result in a general (provisional) report of considerable detail, including the identification of the main factors affecting infant feeding. The details in the interviews, including colourful individual narratives, can be used to highlight probable problem areas in the local situation.
The information in the interviews, together with careful analysis of the interviewing process, also provides a good foundation for training interviewers. Pitfalls in interviewing can be highlighted, special topics illustrated, and the structuring of content areas delineated for the interviewer trainees.
The preliminary, ethnographic data-gathering phase of a project should be seen as the basis for further information gathering and situational assessment. The insights it yields should be tested against data from a larger sample of the population to be served by a programme. The results also need to be considered in conjunction with data on dietary intake, anthropometric characteristics, and other health status indicators. Used in conjunction with other types of information, the insights that are generated by systematic keyinformant interviewing and observation can be very valuable for programme planning, development, and evaluation.
ACKNOWLEDGEMENTS
These guidelines were originally developed in conjunction with a project entitled "The Effect of Food Availability and Infant Feeding Practices on the Nutritional Status of Children 0-23 Months," which is being carried out in northern Cameroon under the auspices of the Centre for Nutrition (DGRST), directed by Dr. Gladys Martin. The ethnographic component of the project was administered by Educational Development Center, through the International Nutrition Communication Service, Ron Israel, Director, with a grant from US/AID (DSAN-C-0209). The assistance of Gladys Martin, Thomas Price, and Pertti J. Pelto in the conceptualization and development of these guidelines is gratefully acknowledged.
