Contents - Previous - Next

The cafeteria is now open: A diverse selection of methods

Despite the tenacious hanging around of the traditional questionnaire (and much like a bad hangover), a quiet revolution in methods has taken place over the past ten years. Although bits and pieces are starting to come out in various forms including mimeographed notes, newsletter, drafts, etc., and even some already published works [5]. This new serving of methods is still poorly diffused (R. Chambers, personal communication, Sept. S. 1983). These innovative methods (actually old methods) find their origin in anthropology, ecology, and geography and are more intensely practiced today in NGOs and locally run "grassroots" research and development projects than by international agricultural centers, government departments, and universities. Below are listed descriptions and examples of a few underutilized techniques which can be added to other appropriate methods which are now well described (ecological transects, field plotting, key informant interviewing, aerial photographs, etc.) [5, 6].

Table 1. Farmers' opinions about disadvantages of agricultural extension programs in District X

Percent of Respondents





Disturbance in the social structure







Traditional varieties may disappear







Ecological balance may tee disturbed







Disturbance in the veil structure







MENTAL MAPS Cognitive models or mental maps are simply fancy anthropological word for how people see and interpret their physical and social environments. Dr. Virginia Sandoval [7] describes a cognitive map as:

"... a geographic representation of the cognitive image pertaining to a certain place or location. In constructing mental maps, people tend to distort proportion and scale. Instead of making faithful reproductions, they exaggerate those areas which are familiar or important to them and downplay those which are not as important or familiar."

For example, in her research in the Philippines, Dr. Sandoval wanted to know how different socioeconomic statuses and gender groups would interpret (mentally map) biotic components of the environment (e.g., plants and arthropods) and major land use options. In one exercise, her informants were asked (12 males and 12 females) to draw the village area. It was emphasized that accuracy was not important and they were to put down on paper impressions which came to mind. Illiterate informants were assisted with labelling while those unfamiliar with the pencil and paper were assisted by the investigator who followed the movement of their hands (working in sand, etc. may overcome this). Dr. Sandoval found significant differences in how the environment was perceived along the lines of gender and social status.

In another case, regular topographical maps were placed by Peruvian researchers in a community room where the communeros gathered in the evening to take coca and aquardiente (local rum)7. After finding that the villagers could not readily read the geographical survey maps, the researchers ask the villagers to draw their own village on paper. Within a very short period of time, complex relationships between zones, households and sectorial rotations were revealed.

Clearly, the way different people perceive their environment depends on the degree of knowledge of that environment. Allowing farmers to create their own maps shows subtle but important spatial relationships that cannot be "seen" by "outsiders." Anthropologists have long known this through in-depth ethnographic studies, but they have never made their methods explicit or understandable to rural development workers. For example, ethnographers record the Alaskan Eskimos recognize and deal with a dozen or so states of snow while outsiders recognize only one or two. Equal knowledge of soil types, parcels, watersheds, etc. are perceived by mountain farmers (and other kinds of "marginal land" farmers). Even in developed countries like the USA, a person from the city will see big pigs and little pigs and many may recognize sex differences (male and female). However to the American farmer, the classifications are subtle: boar, sow, gilt, shoal:, boar pig, weaning pig, runt, not to mention the many breed names which farmers keenly distinguish. In Bhutan, I recently tried to understand from farmers the many names of bovines along a continuum from pure yak to pure cattle. I gave up somewhere about one-fourth along the continuum.

MINIATURE CONSTRUCTIONS OF VILLAGE LANDSCAPES (Spanish: La Maqueta) One of the more exciting activities the Peruvian Pisac group engaged in during the early 1980s was the construction using mud and straw of the village topography. Most high Andean communities are characterized by extremely complex ecologies and farming systems that a one dimensional map fails to reveal, e.g., land use and social relationships. The team, with the communeros, decided to construct a three dimensional model which symbolizes field layouts, irrigation channels, pastures, and the rotation sectors (called turnos or "turns"). In exercises like these, the time and spatial dynamics and folk categories of village agriculture and land use can be understood. Generally, there are seven turns in a village which the villagers can easily detect, but to outsiders the layout was a mystery until explained using the maqueta. The models bring out what is significant to villagers and can stimulate endless hours of discussion of land use, tenure, rotations, disputes, animal ownership, gender assignments, cropping patterns etc.

WORKING FOR FARMERS IN EXCHANGE FOR INFORMATION This was a technique pioneered by John Hatch [8] in his research among corn farmers on the Peruvian coast. The investigator team simply approaches a farmer and asks to exchange labour for learning about a crop or agricultural task. I have found most farmers get a big kick out of this. While the researcher may become the butt end of many jokes, the experience is enriching and engaging for both researcher and farmer. I have found that a full day of harvesting potatoes in the field is physically demanding but extremely rewarding in two ways: a tremendous amount of information can be gleaned and an "insider's" knowledge of the nature of field work can be acquired. This "empathy" is extremely important, since most agricultural scientists today have never done a single day of agricultural labour. Endless variants of this approach are possible. Greg Scott of CIP, who studied potato traders in Peru, claims some of his most valuable information on marketing was obtained by riding potato trucks down the mountain to the central market in the capital city. He helped load the potatoes in the field and experienced the marketing chain right up to the final haggling in the wholesale market of Lima. Variations might include (1) cleaning an irrigation ditch (2) making a mud floor (3) sorting seed (4) herding cattle. Systematically recording the experience in notebooks and analyzing data later is absolutely necessary.

RESEARCHER AS CULTIVATOR OR HERDER A variant of working with farmers is to do it yourself as a farm operator. This is an old technique but has rarely been practiced in recent years by researchers. The method is to acquire a small piece of land and farm yourself. Don Winklemann, now Director of CIMMYT, said in his earliest years in Mexico, as the first economist at CIMMYT, he helped establish his rapport with biological scientists and farmers by actually putting in a crop of corn himself. CIP social scientists in Asia are now learning about indigenous sweet potato cultivation by limiting their inputs and trying to grow their own sweet potato under the supervision of farmers. There is no experience like trying to do it yourself to learn about how farmers make decisions and the problems they face.

USE OF PHOTOGRAPHS In his study of the village of Maracapata in the eastern escarpment of the Peruvian Andes, Dr. Norio Yamamoto of the Osaka Museum of Anthropology had a hard time following farmers' discussions of agricultural calendars and land use on the slopes. As an excellent photographer, Yamamoto took a series of wide angle photos of segments of the village lands and then pasted them together so the complete over view could be obtained. This panoramic photo was then used for discussions with farmers. In another variation, one can show photos of agricultural activities to different groups (male, female, children, elders) and elicit information on their variable perceptions.

EAGLE-EYE PLOTTING AND OBSERVATION A complementary activity, particularly in mountainous areas, is to find a high place above the village lands and plot the fields or watch daily movements of herds and people. Later this can be used with the residents to fill in pertinent information about land use and cropping practices. Quantification of hours of labour, herd sizes, sex and age ratios of humans and herd animals in the fields can be made.

SACK OF SEEDS TECHNIQUE Earlier in this paper, I explained a simple but extremely interesting technique to get at a whole range of topics from family nutrition to marketing practices. Just go to a market place and buy all of the different varieties of the crop or crops that a researcher may be interested in. The seed samples are then taken to a village, or while hiking through a farming area, the seeds are poured out of the sack for farmers to examine and rank. Investigators will be surprised how much information, even quantifiable information, can be obtained in this interactive way. Variations on this technique are to assemble a collection of insect pests pinned in a box or even test tubes or jars with different soil types collected in the study area.

FARMER-DESIGNED EXPERIMENTS Increasingly, agronomists and social scientists at CIP are now letting farmers take over experimentation of the institute's technology from the earliest stages. This includes revolutionary technologies like true botanical seed (TPS) to cut-and-dried tasks such as germplasm evaluation. There is strong evidence that the most successful plant breeders have always used this method; that is, encouraging selection by farmers of successful clones [9]. Another example involves a CIP team working with Indonesian scientists which decided to simply give true potato seed to farmers along with basic concepts and principles and then watch what farmers do. The farmers shortly created many experiments to test alternatives of the potato seed production (direct seeding, transplants, use of tuberlets) (M.J. Potts et al., personal communication, 1987). Farmer advisory boards to design experiments are a variant of this technique.

FARMER FIELD DIAGRAMS This is a farmer-scientist interactive technique that has been described by several authors [10]. This technique is based on the premise that farmers have special conceptual skills that scientists do not have in diagramming important spatial relationships and agricultural flows. These skills do not necessarily involve scientists' tools of pencils and paper but are executed with the tools and materials readily available in a field or a household compound (sticks to draw in sand, pebbles, seed, cut stalks, etc. ) For example, farmers of the Eastern Andes are keenly aware of different ecological floors reaching from the high Andes to the jungle floor covering an additional span from over 4000 metres to 180 metres above sea level. Given a piece of loose soil, they can diagram the different ecological levels, types of crops grown, and give the Quechua names of each level. This sets the stage for hours of interesting discussions about a wide range of topics from politics to plant responses at each ecological level.

TAPPING KNOWLEDGE OF SCHOOL CHILDREN This is a technique used by the CIP post-harvest team on a number of occasions. Research on in-house storage is one of the most difficult tasks for post-harvest researchers due to (1) privacy in the house (seeds are wealth); (2) special gender control (men cannot enter); and (3) suspicion that investigator is a tax assessor. To get around these problems in a Central American country, the team visited a local school, explained the research to the headmaster who in turn sent the children (ages 10-14) home with the homework exercise to draw the kind of storage inside the house. As a team, we explained our purpose to school authorities (and that the information would be guarded). The teacher agreed to make the exercise voluntary and with parental permission. We were amazed at how much information the 30 children brought back on in-house storage of potatoes (design of store, location in house, amount of light and darkness, and physiological conditions of the tubers). In addition, the children were able to elicit information about beliefs and taboos surrounding stored products (e.g., evil-eye).

ELICITING FOLK WEIGHT AND MEASURES The differences between scientists' (metric system) and folk weights and measures have long been a great source of frustration for rural development workers, particularly when using the questionnaire. A simple exercise is to gather items of local units (e.g., basket of potatoes) as defined by the farmers and actually measure to convert to the metric system. In other cases, the conversion is not so simple. For example, farmers in most marginal areas rarely use tons per hectares as a way to measure productivity (they cannot measure tons without scales and they do not know what a hectare unit of land is). Instead, they frequently use multiplication rates (e.g., one sack of seed potatoes yields six sacks of potatoes). The simple way to acquire this information is through a seed multiplication game on a simple board or in the sand where you define the zones (e.g., irrigated and upland). A single tuber potato may represent a sack of a given variety and the question to the farmer is how many tubers can be produced from this one.

CASHING IN ON MENTAL BANKS This is a new concept being developed by Dr. Virginia Sandoval of CIP (Philippines). The idea is that "germplasm" banks (such as IRRI for rice, CIP for potatoes or ICISAT for sorghum) have an analog in "mental" banks possessed by farmers who have long cultivated, nurtured, and utilized those varieties. The importance of "mental" banks is underscored in two ways: (1) many germplasm gene banks contain materials that can be found only in the "mental" banks of people over 70 years of age (in the case of rice in the Philippines). Samples of the vanished varieties taken from gene banks' materials can be discussed with the people who once used them. After the older generation passes, all existing knowledge about varieties will be lost; (2) knowledge in the form "mental" banks can be linked with ongoing agricultural research (such as classification and evaluation of recently collected germplasm). The specific techniques for drawing on mental banks will vary from informal interviewing or discussions while examining objects of concern (e.g., seed). Mental banks can be also approached through a simple opening question "What varieties of rice (or potato, sorghum, etc.) did your father and mother grow when you were a child? [11]. When you were a young person of 15? And when did you first start farming on your own land?" Exploring mental banks at different time periods can give a long-term dynamic view of specific germplasm problems.

OLD FASHIONED ETHNOGRAPHY Another out-of-vogue method (along with the questionnaire) is traditional ethnography which has been described as "10 years of research and 10 years write-up." However, a great deal of anthropological research has always been rapid (although the ideal for the student was once to reside in a village for a year or more). Even Robert Chambers who formerly criticized traditional ethnography is now talking about "camping" in the village, one step back toward "living" in the village. In fact, there is no real substitute to spending time with people. Only then can you truly penetrate the "deep structures" of their lives and livelihoods.

The 13 "new" methods listed above do not exhaust the new menu but are merely illustrative. In using these and other innovative methods, the following cautionary notes should help the user keep the new methodology in perspective:

1. These methods are always informal, creative, flexible and interactive. They are to be embellished as the researcher and researched see fit. As a result, no cook book on "how to do it" step by step should be published, although guidelines and examples are urgently needed.

2. There is no better training than experience. Hands on training courses are needed. The only way to learn the new methodology is to use it.

3 The new methodology may not always be quick and cheap. Some of these procedures take time.

4. The methods can be used in clusters and each should flow easily into the other. There is no rigid sequencing of methods but the context and natural flow will lead the investigator to vary the order of techniques.

5. The implementor of the new methodology must be careful not to reduce the interaction with farmers to a kind of insulting silliness. The cultural context must be understood as to what is permissible and honorable. If an investigator uses any of these participatory techniques because he thinks farmers have a lower level of ability for conceptualization and therefore must be treated like children, then the investigator has obviously missed the point.

6. Those bold enough to use the new methods can expect to be ridiculed by normal scientists. Already jokes are circulating in formal science circles about those who play funny little shell games with farmers and believe in finger scales that farmers use to measure animals and children.

7. While these techniques are meant for action research, however, it is my contention that they are also valuable in and of themselves for any serious researcher (pure or applied) who wishes to understand Third World farmers. Quantified data as well as descriptive, anecdotal data can be gathered by these methods. If necessary, scientific-looking charts, pie charts and tables can be generated from these informal methods if "numbers" proves to be the most influential form of communication for convincing agricultural scientists or policymakers.


In this paper, I have been strongly critical of quantitative field methods used in rural and agricultural development research. I believe my criticisms are well taken, but it is equally true that those academic fields - particularly anthropology and geography - which have pioneered the methods described in this paper have not made any attempt to diffuse them for wider use beyond their narrow academic interests. Of equal interest for rural development workers would be the social theory (symbolic interactionism, grounded theory, decision-making trees, etc.) which underlies the new methodology. Volumes such as Applied Qualitative Research [12] should be distilled and reworked to apply more directly to agricultural and rural development projects. The point is that there is a great deal of value in the academic literature which is not being utilized in the field.

Finally, the revolution in methods will only succeed if a simultaneous revolution in thinking about farming and rural life takes place among development researchers [13]. Not only will rural development researchers have to become "insiders" as much as possible (and these techniques encourage the inside view), but agriculture itself should be understood as "performance" in the parlance of Paul Richards [14]. Third world farming has almost never been a predetermined plan concocted by cultivators who select this technology or that technology off the shelf to meet a "constraint." Instead, farming may indeed be perceived by farmers as having an imagined "ideal" but due to totally unpredicted events that occur along the way (selling the cow to bury grandmother), the act of farming becomes a fluid, amorphous drama with no end point. Farmers, of course, will tell us that their performance can and should be improved. Methods that involve rural development workers' analysis of that ongoing drama are destined to bring insights (and technological solutions) never creamed up by normal science. Farmers already know this. Their eyes, their laughter, their hand gestures, and a beautiful enthusiasm only found among rural peoples are clues as to whether we are coming closer to understanding the performance. Convincing established, normal science that the revolution is upon us will be a much more difficult and serious undertaking.


1 Workshop on "Methodological issues facing social scientists in applied crop and farming systems research." CIMMYT; Mexico (April 1-3, 1980).

2 The participant list, if I remember correctly, read something like today's Who's; Who of international agricultural economists: Don Winklemann, Peter Hildebrand, David Norma, Jock Anderson, Mike Collinson, Derek Byerlee, John Flinn, Larry Harrington, Doug Horton to mention only a few.

3 In fact, the Art of the Informal Survey was rejected by several journals, including Human Organization (which later accepted a stilted academic re-write) and Agricultural Administration. Reason: Not "Scientific." Meanwhile, as Robert Chambers later informed me, the paper was being photocopied on hundreds of worn out, wired together copy machines in dusty agricultural extension and research offices throughout his beat. Practitioners in the field were hungry for such relevant methods.

4 I am not including in this category survey of market prices.

5 In 1967, I was an exchange sociology student at the University of the Philippines, Los Baños. The only respectable kind of survey at that time was the elaborate Everet Rogers styled questionnaire.

6 I prefer not to cite the reference from which this question is taken.


1. Rhoades R. Notes on the art of the informal survey. Paper presented at the methodological issues facing social scientists in applied crop and farming systems research. Mexico: CIMMYT, April 1-3, 1980.

2. Horton D. Social scientists in agricultural research: lessons from the Mantaro Valley Project, Peru. Ottawa: IDRC, 1984. 67p.

3. Rhoades R. Traditional potato production and farmer selection of varieties in eastern Nepal. Potatoes in food systems research series report no. 2. Lima: CIP, 1985.

4. Diamond D. Soft sciences are often harder than hard sciences. Discovery 1987; August: 34-39.

5. Khon Kaen University. Rapid rural appraisal, proceedings of the 1985 international conference. Thailand: Khon Kaen University, 1987.

6. Rhoades R. The art of the informal agricultural survey. Social science department training document 1982-2. Lima: CIP, 1982.

7. Sandoval V. Philippine rural cultivators in transition: operational reality and cognized models in agricultural decision making. PhD dissertation. Investigating Lexington, Kentucky: University of Kentucky, 1987.

8. Hatch J. The corn farmers of Motupe: a study of traditional farming practices in northern coastal Peru. Land tenure centre monograph no. 1. Madison: University of Wisconsin, 1976.

9. Prain G. The friendly potato: farmer selection of potato varieties for all occasions. In: Moock J, Rhoades R. (eds.), Farmer knowledge and sustainability. Ithaca, NY: Cornell University Press, in press.

10. Lightfoot C, Axinn N. Singh P. Bottrall A, Conway G. Training resource book for agro-ecosystem mapping. India: International Rice Research Institute and Food Foundation, 1989.

11. Box L. Virgilios theorem: a method for adaptive research. In: Chambers R. Pacey A, Thrupp LA (eds.), Farmer first. London: Intermediate Technology Publications, 1989; 61-67.

12. Walker R. Applied qualitative research. London: Gower, 1983.

13. Rhoades R. Booth R. Farmer-back-to-farmer: a model for generating acceptable agricultural technology. Agric Admin 1982; 2: 127-137.

14. Richards P. Agriculture as performance. In: Chambers R. Pacey A, Thrupp LA (eds.), Farmer first. London: Intermediate Technology Publications, 1989; 39-43.

Contents - Previous - Next