Contents - Previous - Next
Bongaarts J. Reining P. Way P. Conant F. The relationship between male circumcision and HIV infection in African populations. AIDS 1989; 3(6): 373-377.
Casley D, Kumar K. The collection, analysis, and use of monitoring and evaluation data. Baltimore, MD: Johns Hopkins Press for the World Bank, 1989.
Cernea MM, Tepping B. A system for monitoring and evaluating agricultural extension projects. Washington DC: World Bank Staff Working Paper No. 272, 1977.
Cernea MM, ed. Putting people first. New York, London: Oxford Univ. Press, 1985.
Cernea MM. The production of a social methodology. In: Eddy EM, Patridge W. eds. Applied anthropology in America. New York: Columbia University Press, 1989:237-262.
Cernea MM. Social science knowledge for development interventions. Cambridge: Harvard University, Harvard Institute for International Development, Development discussion paper no. 334, 1990.
Chambers, R. Participatory shortcut methods of gathering social information for projects. In: Cernea MM, ed. Putting people first. Sociological variables in development projects. New York: Oxford University Press, 1991:515-537. Note: See also Chamber's paper in the 1st ed. of this vol. (1985).
Chambers R. Pacey A, Thrupp L, eds. Farmer first: farmer innovation and agricultural research. London: Intermediate Technology Publications, 1989.
Green EC. Anthropology in the context of a water-borne disease control project. In: Green EC, ed. Practicing development anthropology. Boulder and London: Westview Press, 1986.
IIED: RRA Notes, No. 1 (June 1988) to No. 9 (August 1990). (This series publishes an excellent collection of notes, articles, descriptions, news, etc. about rapid rural appraisal, with the aim of sharing experiences and methods among practitioners of RRA throughout the world. Published under the Sustainable Agriculture programme of the International Institute for Environment and Development (IIED), London.
Molnar A. Community forestry rapid appraisal. A review paper. Rome: Food and Agriculture Organization of the United Nations, 1989.
Murphy J. Monitoring and evaluation of extension projects. Technical paper. Washington DC: World Bank, 1988.
Pedersen D. Qualitative and quantitative: two styles of viewing the world. In: Scrimshaw NS, Gleason GR, eds. Rapid assessment procedures: qualitative methodologies for planning and evaluation of health related programmes. Boston, MA: International Nutrition Foundation for Developing Countries (INFDC), 1992.
Ruttan, Vernon W. Cultural endowments and economic development: what can we learn from anthropology? Economic Development and Cultural Change, 1988; 36(3): 247-271.
Salmen LF. Beneficiary assessment: improving the design and implementation of development projects. Eval Rev, 1989; 13(3): 273-291.
Scrimshaw SCM, Hurtado E. Rapid assessment procedures for nutrition and primary health care. Anthropological approaches to improving programme effectiveness. Los Angeles: UCLA Latin American Center, 1987.
Scrimshaw S. Adaptation of anthropological methodologies to rapid evaluation of programmes of nutrition and primary health care. In: Scrimshaw NSS, Gleason GR, eds. Rapid assessment procedures: qualitative methodologies for planning and evaluation of health related programmes. Boston, MA: International Nutrition Foundation for Developing Countries (INFDC), 1992:
TN. On studying socio-economic change in rural India. In: Bardhan
P. ed. Conversations between economists and anthropologists.
Oxford: Oxford University Press, 1989.
Annett H. Rifkin S. Guidelines for rapid appraisal to assess community health needs. A focus on the health improvements for low-income urban areas. The Swedish International Development Agency (SIDA)/Swedish Agency for Research Cooperation with Developing Countries (SAREC), 1989.
Appadurai A. Small-scale techniques and large-scale objectives. In: Bardhan P. ed. Conversations between economists and anthropologists. Oxford: Oxford University Press, 1989.
Cicourel, A. V. On the microtranslation of macrosociology. Am J Soc 1981; 86: 984-1014.
Epstein S. A manual for culturally adaptive market research (CMR) in the development process. East Sussex: NWAL, 1988.
Foster GM, Scudder E, Colson, Kemper RV, eds. Long-Term research in social anthropology. New York: Academic Press, 1979.
Franzel S. Crawford E. Comparing formal and informal survey techniques for farming systems research: a case study from Kenya. Agric Admin 1987; 27(1): 13-33.
Hildebrand P. ed. Perspectives on farming systems research and extension. Boulder: Lynne Rienner Publishers, 1986.
Khon Kaen University. Proceedings from the 1985 international conference on rapid rural appraisal. Rural systems and farming systems research projects. Thailand: Khon Kaen University, 1987.
Knorr-Cetina K, Cicourel AV. Advances in social theory and methodology. Boston and London: Routledge and Kegan Paul, 1981.
Kumar K. Rapid low cost data collection methods for AID. AID program design and evaluation methodology. Report No. 10. Washington, DC: USAID, December 1987.
Leach E. An anthropologist's reflections on a social survey. In: Jogmans DG, Gutkind PCW, eds. Anthropologists in the field. New York: Humanities Press, 1967.
McCracken J, Pretty J, Conway G. An introduction to rapid rural appraisal for agricultural development. London: International Institute for Environment and Development (IIED), 1988.
Merton RK, Riske M, Kendall PL. The focused interview. A manual of problems and procedures. Glencoe: The Free Press, 1956.
Nath Roy R. Getting people to plan: approaches to participatory planning. Madras: Bay of Bengal Programme, FAO, 1987.
Oquist, P. The epistemology of action research. Acta Sociologica, 1978; 21(2): 143-163.
Patton MQ. Qualitative evaluation methods. Beverly Hills and London: Sage Publications, 1980.
Pelto PJ. Anthropological research. The structure of inquiry. 2nd edition. New York, Evanston and London: Harper and Row Publishers, 1978.
Rudqvist A. Planning, diagnostic studies and popular participation. University of Stockholm: Development Studies Unit, Department of Anthropology, 1990.
Learning from the field, a guide from experience. Beverly Hills
and London: Sage Publications, 1984.
The UNU sixteen country study
Additional applications of the RAP guidelines
By Susan C. M. Scrimshaw
Susan Scrimshaw, Ph.D., is Associate Dean of the School for Public Health at UCLA.
This paper reviews the history and current status of Rapid Assessment Procedures from the perspective of one of the most well known pioneers and leaders of the methodological approach. Scrimshaw describes well both the accomplishments of RAP to date and its limitations. Several issues touched on in this paper, such as training, were the topic of later conference panels. - Eds.
PROGRAMMES FOR HEALTH and development have seldom waited for lengthy and careful scientific studies to guide their development or revision. The traditional approaches of anthropology, sociology and related fields such as demography have proved too expensive, time-consuming and cumbersome for all but a few well funded or longitudinal projects. Data generated from traditional applications of these fields may be picked up and applied to projected and ongoing programmes in some instances, but, as a rule, programmes are evaluated by quick counts of vaccinations given or the impressions of a supervisor during a brief visit. The stacks of paperwork imposed on most health programmes by the bureaucracies of the world are seldom processed, and their generation takes valuable staff time away from patients.
These practices are supported by the orthodoxy of many social scientists, the lack of communication between social scientists and providers and planners of health care, and the assumption made by programme planners and providers that they know what people need and how to provide it. But, these practices have also been challenged by applied social scientists, including applied anthropologists and sociologists and rural sociologists. The core of their criticism is expressed in descriptions of failed or troubled projects, where the lack of attention to cultural and community norms, concerns, and beliefs can be linked to many of the problems encountered by the project [1-10].
In addition to the more formal, recorded efforts discussed above, practitioners in these fields have often given quick advice based on their years of experience with cultures and programmes. Often, the practical questions asked by programme planners and health care providers could be answered with information derived from examining the existing anthropological and rural sociological literature and suggesting common sense solutions or by a few days or weeks of exploratory observation and interviewing [11-13]. This work, and the methods used, was unlikely to be shared with the originating profession since it did not constitute formal research.
work on formalizing methodologies for rapid assessment using
qualitative data collection techniques came from rural sociology,
and has been called "Rapid Rural Appraisal" by its
developers [11, 14-17]. While these methods were being applied
primarily to agriculture and rural development, anthropologists
working in public health began to formalize their own
"shortcuts" [12, 13, 18, 19]. It was an attempt to
systematize this practical approach to programme planning and
evaluation which led to the United Nations University Sixteen
Country Study and to the development of the RAP field manuals.
Epidemiologist Ralph Frerichs perceived the same need, and took
advantage of laptop computer technology to make rapid
epidemiological surveys possible .
The need for rapid assessment based on anthropological strategies emerged because UNU leaders were discouraged and frustrated with the expense, the time consumed, and the relatively poor validity (accuracy) of standard survey techniques to assess health care. In their experience, some surveys were even conducted but never analyzed. As discussed above, they noted that much programme planning and evaluation is based on statements from workers in health posts and clinics of services they said they provided, or on brief visits to health service sites by more senior officials, who might not get the real picture in a few formal hours. UNU researchers were concerned with what services people were actually obtaining and using, with people's perceived needs and concerns, and with their use of alternative health resources (private biomedical services, pharmacies, and indigenous medicines and practitioners).
The objective of the sixteen country health seeking behaviour studies was to assess nutrition and primary health care programmes from the household perspective, and to do so as rapidly and accurately as possible. Researchers were to select communities where the government officials felt there was a good primary health care programme in place, and to learn from the household members their perceptions of health and illness, their resources for preventing and treating illness, and their actual use of these resources. Thus, the programmes were to be evaluated from the household up rather than from the health programme down.
The United Nations University planners were thus looking for an alternative to the usual approaches to programme evaluation. In general, large scale surveys have been regarded as the method of choice for formal programme evaluation. Quantitative surveys permit the collection of data from large numbers of people in standardized ways, enabling comparison between communities, countries and time periods. Alone, however, they are often insufficient in providing the type of in-depth information required to understand the complexity of human behaviour and to formulate prevention, and control strategies and programmes. As discussed elsewhere , qualitative information needs to be generated on the same topics in a way which permits a more detailed, a more rapid, and often, a more accurate understanding of the underlying social and cultural characteristics influencing or associated with specific patterns of behaviour.
In fact, programme evaluation in actual practice, usually does not involve large scale surveys. Instead, it takes the form of field trips by national or international experts who visit clinics, talk to staff, and look at records for one or a few days. Frequently, major funding and related programmatic decisions are made on the basis of such visits, although they are not considered "research", nor should they be. This "quick and dirty" programme assessment is usually done when time and money are not available for a large scale survey, and is often considered the only other alternative. The problems with this type of evaluation are that it is idiosyncratic to the evaluators rather than systematic, and it is prone to error for many reasons. It is superficial, people are trying to impress the evaluators, the visitors may not be familiar with the country, the language, the class, tribe or region, and it does not probe for the reasons for a programme's success or failure.
The application of anthropology to programme planning and evaluation assumes that there are other important tools for understanding human behaviour in addition to the large survey and the "quick and dirty" field trip. In fact, everything quick is not necessarily dirty, and time does not insure cleanliness. Among these tools is the application of the traditional anthropological techniques of observation, participant observation, formal inter view, informal interview, conversation, and group discussion (now called focus groups) to the evaluation of health programmes.
Anthropological research attempts to understand social phenomena from an insider perspective rather than by imposing an investigative framework from the outside. It acknowledges the unexpected and places such findings in their appropriate cultural context. It deals with cultural nuances such as the discrepancy between ideal and real behaviour by understanding the underlying mores of the community and by relying heavily on observation as well as other multiple sources of information. For example, some cultures maintain beliefs that the male head of household makes all major family decisions. Consequently women's roles in illness diagnosis, food production, and treatment seeking might be hidden and not readily acknowledged in questionnaire interview situations. Such situations call for research approaches that explore behaviour using discreet and sensitive techniques that encourage openness and the sharing of more information.
In anthropological work it is stressed that nothing is taken for granted. The researchers make every effort to avoid placing their own biases on the collection and interpretation of information. During this task, observations become as important as information obtained during interviews in learning about attitudes, knowledge and behaviours regarding illness. These observations and the statements from informants comprise the data that are recorded as completely and as accurately as possible.
The traditional anthropological approach involves one person or a small team in a research site for at least a year. This is done in part to take into account the variations in people's lifestyles with the changes in seasons, activities, available food, and so on. The scope and complexity of data collected are extensive. Also, the anthropologist often needs time to learn a language or dialect and learn enough about the culture to provide a context for questions and observations.
For the application of anthropology to the evaluation of nutrition and primary health care programmes, it was necessary to alter this traditional pattern. One year was far too long if the information was to be useful in health programme improvement. Instead, researchers already knowledgeable in the culture could base smaller rapid studies on their earlier experience. In addition, it was considered important to develop local expertise with indigenous researchers who would remain available to their countries and regions and could train others.
For the United Nations University Sixteen Country Study, an abbreviated plan was outlined that depended on two factors:
1. Researchers would be individuals already familiar with the language and the culture. The researcher, at least the senior researcher guiding the team, would already have conducted research in the culture and would have a good understanding of customs, dialect, and behavioural meanings.
2. Researchers would work with a limited list of objectives, or data collection guidelines, tailored to the programme evaluation goals. This would permit focused work which could be carried out relatively quickly.
The group of UNU supported researchers was selected to meet the first criterion. Next, a common methodology and set of data collection guidelines were needed. These were developed by the group at its first meeting in Geneva in 1983, on the basis of earlier work in Central America by E. Hurtado and S. Scrimshaw. The guidelines were grouped by community related data and household based data. Later, guidelines for data to be collected at health resources were added. The RAP guidelines were designed to allow anthropologists and other social scientists with relevant training to spend approximately six weeks in a community where the government believed an effective primary health care programme was in place, and to obtain the household and community perspective on those services.
As demonstrated by the example in Table 1, these guidelines were like shopping lists rather than questionnaires. Phrasing was deliberately brief to encourage the development of local wording. These guidelines covered topics such as availability, use, and personal opinions of both traditional and western biomedical health resources, definitions of disease and illness, and common illnesses in children and their treatment. Each researcher then took the lists and made culturally appropriate modifications. The data were collected using the following techniques:
Informal interviewing: Open-ended questions are asked and recorded about specific topics following a general outline and allowing additional subjects to be incorporated as they arise.
Conversations: Informal conversations with informants or with small groups are incorporated in the data.
Observation: Careful documentation of observed events and behaviours provides valuable nonverbal clues as to what is actually occurring.
Participant observation: Participation in and observation of the daily socio-cultural context of a household or community.
Focus groups: Small homogenous groups are gathered for group discussions of appropriate research topics.
Collection of data from secondary sources: Previously published and unpublished research, government and community records, and health services records.
Some structured questions used for inventories and demographic information.
Because each researcher had the freedom to make culturally appropriate changes and to incorporate additional information important to the local situation, the results do not have the "cookie cutter" sameness of a c cross-national survey. On the other hand, because the perceived meaning of even a well translated question may vary from culture to culture, or even within cultures, cross-national surveys may not in fact have the hoped for validity and reliability.
What emerges from the approach used by the researchers in those early studies are descriptions of beliefs and behaviours on the same topics across 514 households in 42 communities in 16 countries. The wording of the questions was not identical for each person studied, but the accuracy and appropriateness of the information has been verified by triangulation.
Table 1. Definitions of Health and Illness (open-ended)
Triangulation involves cross-checking data for accuracy through the use of repeated questions, discussion and actual observation [22, 23]. The question guidelines not only help to focus the data collection, they also speed up and simplify analysis. Data can quickly be grouped by topic and conclusions drawn. One analytic and reporting device which proved to be particularly successful was to diagram contrasting views of a situation between recipient and providers. For example, taxonomies showing illness classification within the community in contrast to that of the primary health care workers were helpful in explicating obstacles to programme use [24, 25]. For example, when malaria and leishmaniasis are perceived as the same illness, as for some Kenyan families , it is difficult to introduce vector and disease control strategies unless that perception is understood and addressed.
One issue which concerned the researchers was that of sampling. It was agreed that households with children under five would be studied, since the goal was to evaluate nutrition and primary health care programmes. It was also agreed that, for the most part, families in need of such services would be studied. This meant a focus on poor and rural households. The group agreed to try random sampling, but it was only possible in nine countries. In many areas it is not feasible because data needed for sample selection are unavailable. Mapping and listing all the households in a community could take longer than the time allocated for data collection itself. Instead, purposive or opportunistic sampling are more feasible. Researchers found that their concerns for representativeness could be honored by a strong awareness of what was typical or deviant for the culture. Also, families could be added to the sample if more seemed necessary because of a wide variability in responses.
initial projects in sixteen countries did reveal some surprises.
As summarized in Table 2, the patterns of identifying illness and
seeking treatment, and the obstacles to the use of primary health
care programmes were remarkably similar across countries. This
appears to reflect consistencies in the Western biomedical
systems and primary health care approaches as they interface with
families who have their own intricate social patterns and belief
systems. More detailed reports from the sixteen countries are
compiled in S. Scrimshaw, K. Mitzner and N. Scrimshaw, eds. 
Since those first sixteen studies, many have been added and the original researchers have trained others in their countries and regions. The guidelines were revised on the basis of the initial field experiences, and now are available in at least six languages [12, 24].
Table 2. Summary of Sixteen Country Study Using the RAP Field Guide
COUNTRIES: Kenya, Nigeria, Zaire, Haiti, Nicaragua, Guatemala, Costa Rica, Peru, Colombia, Chile, Nepal, Bangladesh, Korea, Pakistan, Philippines, Thailand
RESEARCHERS: Mostly anthropologists. 1 MD, 1 economist, 1 sociologist
TOTAL COMMUNITIES: 46 (But 17 in Pakistan. Without Pakistan, then average is 2 per country) Most communities were rural. Some were low socioeconomic urban areas.
TOTAL HOUSEHOLDS STUDIED: 514 (Pakistan did not distort this since there were fewer households studied per village there)
TIME IN FIELD: Most studies took 2 to 4 months, but the range was 1 to 12 months. Studies were conducted during the period ranging from 1983-1985.
SAMPLE: Random for 9 countries; otherwise, purposive, total, or opportunistic
Universal: (Found in all the communities)
Policy Regarding Indigenous Practitioners (as reported by community members)
Availability of Modern Preventive Care (as reported by community members)
Availability of Modern Curative Care (as reported by community members)
RAP manuals have been developed for Epilepsy . An adaptation for the study of HIV/AIDS is also forthcoming [25, 29]. Simpler versions have been used to help community health workers evaluate their constituents. Programme planners have used the tools for evaluation visits. More has been Programme planners have used the tools for evaluation visits. More has been learned about the speed which is possible when the questions are few and specific. The original sixteen country studies were designed to take six weeks in the field, six for analysis. It is now clear that some problems take longer, but that waiting room procedures can be improved after a day or two of study. In Kenya, the head of a condom distribution programme from a neighbouring country tried to buy condoms in all three drugstores in a small Kenyan town. The sellers couldn't or wouldn't tell her how to use them. The implications are clear: If this occurred in only a few more pharmacies the programme would need to address the issue of helping pharmacists to become better sources of education and information for condom use.
In some cultures, application of the RAP is more difficult. Even for researchers who know the language and the culture, and who are working in their own country, communities may be wary of outsiders. But this problem is no different for other methods. It may be better to know where uncertainties lie than to assume answers given to assuage an international or local official are truthful, in a culture where the truth is not a stranger's business.
As the use of anthropological methodologies for rapid assessment has grown, several concerns have emerged. First, many anthropologists are concerned that people will think the method is anthropology. They argue, correctly, that anthropology involves a large body of theory which guides research, and that anthropological research is lengthy and comprehensive.
RAP does not, by any means, replace traditional anthropology. It must be seen as an additional method, the application of some of the tools employed by anthropologists, and the application of the concept of looking at cultures from within as well as from the outside. Rather than "replacing" traditional anthropology, the RAP approach can create opportunities for anthropologists to apply their craft and experience to brief problem-solving efforts. This may indicate the need for depth work rather than detracting from it.
It has also been mentioned that the approaches outlined in the RAP guidelines are not new, that the techniques have been used by anthropologists, rural sociologists, and others. As mentioned in the introduction, other disciplines and individuals have indeed worked with these methods, and applied anthropologists have used them for many years. Because of the need for a common methodological handbook, the researchers involved -in the Sixteen Country Study project packaged these methods in a particularly accessible way. That packaging has increased the visibility and use of the approach.
There is also the fear that people will think this approach is a panacea, which will solve all our questions about how to design and evaluate programmes. In fact, it is only one approach among many. It has advantages and drawbacks, which are discussed throughout this volume. It works well in complement with other methods, and can also be used alone, but it does not imply that anthropologists have all the answers. They are part of a team trying to solve a very complex set of problems.
In sum, the approach is the product of the development and refinement of specific data collection guidelines designed to be used by culturally knowledgeable researchers employing anthropological techniques. This application of anthropology to the rapid assessment of nutrition and primary health care programmes must be regarded as adding one more tool to the toolbox available for programme development and evaluation research. It does not replace the other tools, but complements them. Everything does not become a nail because a hammer is available. Sometimes work can be done primarily with a hammer, other times several tools are used together or in succession. The rapid anthropological assessment methods described in this volume are being used alone, together with other methods, and preceding or following other methods. They have common themes with approaches used in rural sociology, marketing, applied anthropology and related fields. They strive to be rapid, practical for problem solving, and to provide accurate data through the use of multiple information sources. Most of these approaches use data collection "shopping lists" or guidelines. All face similar issues:
Convincing others of the validity and reliability of these approaches.
Finding or training skilled evaluators and field workers.
Interpreting results in ways understandable to the consumers, the community and the programme planners and providers.
Recognizing the limitations of the approach.
Contents - Previous - Next