Indigenous Socio-Economic / Demographic Database Description

The indigenous socio-economic/demographic (SED) database contains a wealth of information. Below is a description of the data collection effort. We have not yet made any variable lists of descriptive statistics available for public dissemination. Those should be available in the future.

I. Household and Community Surveys

In 2001, household and community surveys were administered in 36 indigenous communities. Communities studied are from five indigenous populations in the northern Ecuadorian Amazon: the Quichua, Shuar, Huaorani, Cofán, and Secoya. The surveys, when made available, will be on the Indigenous Questionnaires & Manuals page.

II. Brief Description of the Samples for the Household Survey and the Community Survey

A two-stage sampling procedure was used to select, first, communities and then households within selected communities. At the community level, selection was based upon controlled sampling (see Kish, 1965:494; Goodman and Kish, 1975:351) in order to select communities that appeared to represent a range of different conditions in terms of location/access to roads and towns, biophysical characteristics, and population size and density. The number of communities sampled from each indigenous group was determined in relation to the size of the different indigenous populations. Therefore Quichua and Shuar communities make up over half of the sample, with the other indigenous groups--Huaorani, Cofán, and Secoya--sampled in smaller numbers to capture different degrees of exposure to colonization as well as different cultural characteristics.

Within sampled communities, households were sampled according to two rules. In most of the 36 communities, due to the small number of households per community, all households were interviewed. In larger communities, a maximum of 22 households per community was determined sufficient to normally yield 20 completed households. A sample frame was prepared by the field supervisor and the community leaders showing the location of each occupied dwelling, and 22 households were randomly selected from this map-listing. This differential sampling procedure leads to different probabilities of selection for households from large versus small communities. Selection weights were calculated for each household and can be used in the analysis.

The original sample included a total of 565 households, including the eight communities already covered in the ethnographic phase, plus an additional 28 communities covered in the survey. The number of completed household interviews (both head and spouse) is 480¹. The refusal rate was lower than 10%, which is low for a survey of indigenous communities that have traditionally resisted such research efforts. This was due to good relationships established with indigenous federation leaders during the ethnographic fieldwork, prior visits of senior project Ecuadorian and U.S. investigators to many of the prospective survey communities beforehand, the cultural sensitivity and interviewing skills of the Ecuadorian project coordinator and field supervisors, and the receptivity of most indigenous community leaders and residents.

The questionnaires used for data collection are similar to the instruments used for interviews with colonists in the study area in 1990 and 1999 (see description of NASA project, on this website). This similarity allows for comparison and contrast of important factors influencing land and resource use of colonist and indigenous populations in the northern Ecuadorian Amazon study region. Thus the household head’s questionnaire covered household location, origin and migration of head, land tenure and use, production and sale of crops, any raising of cattle or other large animals, off-farm employment, hunting and fishing, technical assistance and credit, perceptions of environmental contamination, and attitudes and aspirations for children’s education and permanence in the community. Besides covering the same topics in connection with migration origins, the environment and aspirations, the spouse’s questionnaire included a household roster listing all members of the household by age, sex, education, marital status, etc., out-migration from the household, household assets, and fertility, mortality and health. If either the female or male head of household was absent due to death, divorce, or migration, both questionnaires were implemented with the person available to ensure complete data collection for each household.

The community level survey was implemented with village leaders in each community. The questionnaire covered a variety of topics, including: land title history, hunting and fishing resources, population (number of households as well as in- and out-migration), community infrastructure, location and access to external facilities (markets, health centers, secondary schools, etc.), contact with other communities, and contact with outside organizations and individuals.

References

Goodman, R. and L. Kish. 1975. “Controlled selection: a technique in probability sampling.” Journal of the American Statistical Association 45: 350-372.

Kish, L. 1965. Survey Sampling. New York: Wiley and Sons.

¹ Of the original sample of 565 households, 509 and 506 head and spouse interviews respectively were completed, for a total of 531 households that completed at least one of the two questionnaires.