Issues of scale for environmental indicators

Integrated pest management is promoted in coffee plantations to control pests and disease in a manner less harmful to the environment than the use of pesticides alone. The rate of adoption of these practices is variable, possibly influenced by different social, economic, environmental and institutional factors. This was explored by fitting standard non-linear curves to uptake data for each of four chosen integrated pest management (IPM) recommendations for control of coffee berry borer, Hypothenemus hampei (Ferrari), in Colombian coffee. Logistic curves were shown to be the most efficient for all four recommendations. Logistic regression analysis was then used to determine the impact of different factors upon the uptake of the recommendations singly and in combination. Comparisons are made between the results of the analyses to confirm the choice of the single or combined datasets and the reliability of the models. The results showed that different factors affected the adoption processes at different stages in time for the different recommendations, both when used singly and when in combination. A link was demonstrated between level of education, wealth of the farmer and choice of recommendation, poorer farmers choosing recommendations that did not require a large financial outlay or which required a high level of technological skill. Suggestions for more controlled studies of adoption are presented.