On-farm trials: A survey of methods

Abstract

After more than a decade of project work on Farming Systems Research, much confusion and disagreement surrounds the methodology for on-farm experiments. This paper reports the results of a survey of 41 practitioners who were asked to report on the methods and procedures they were using in their projects, and the reasons for success or failure. Responses suggest that projects based on formal, complex, researcher-designed experiments ran into a maze of problems including logistical support, analytical needs, interdisciplinary compromise, and farmer participation. Most importantly, this conventional approach to farming systems research fails to incorporate the experimential knowledge of the farmer in the research design. Thus, many if not most farming systems research projects fail to provide useful information to farmers or to station-based researchers. On the other hand, those practitioners who do report success indicate that experiments which exploit indigenous knowledge to define research questions, and employ many farmers to conduct simple, flexible trials can provide the information needed to develop technologies relevant to resource-poor farmers.