Evaluating indices of traditional ecological knowledge: a methodological contribution
1 ICREA and ICTA, Universitat Autònoma de Barcelona, 08193 Bellaterra, Barcelona, Spain and Sustainable International Development Program, Brandeis University, Waltham, MA 02454, USA
2 Sustainable International Development Program, Heller School for Social Policy and Management, Brandeis University, Waltham, MA 02454, USA
3 Department of Anthropology, Northwestern University, Evanston, IL 60208, USA
Journal of Ethnobiology and Ethnomedicine 2006, 2:21 doi:10.1186/1746-4269-2-21Published: 25 April 2006
New quantitative methods to collect and analyze data have produced novel findings in ethnobiology. A common application of quantitative methods in ethnobiology is to assess the traditional ecological knowledge of individuals. Few studies have addressed reliability of indices of traditional ecological knowledge constructed with different quantitative methods.
We assessed the associations among eight indices of traditional ecological knowledge from data collected from 650 native Amazonians. We computed Spearman correlations, Chronbach's alpha, and principal components factor analysis for the eight indices.
We found that indices derived from different raw data were weakly correlated (rho<0.5), whereas indices derived from the same raw data were highly correlated (rho>0.5; p < 0.001). We also found a relatively high internal consistency across data from the eight indices (Chronbach's alpha = 0.78). Last, results from a principal components factor analysis of the eight indices suggest that the eight indices were positively related, although the association was low when considering only the first factor.
A possible explanation for the relatively low correlation between indices derived from different raw data, but relatively high internal consistency of the eight indices is that the methods capture different aspects of an individual's traditional ecological knowledge. To develop a reliable measure of traditional ecological knowledge, researchers should collect raw data using a variety of methods and then generate an aggregated measure that contains data from the various components of traditional ecological knowledge. Failure to do this will hinder cross-cultural comparisons.