An illustrated review of some farmer participatory research techniques

A review of farmer participatory research is given highlighting its variety and wide range of terminologies. Two types of methodology are discussed, rapid rural appraisals (RRA) and participatory rural appraisals (PRA). Some methods included in the PRA group generate quantitative and qualitative data suitable for statistical analysis. The need to ensure rigor in these studies is emphasized, but this is often shunned by researchers in the belief that their studies are simple and require no more than an elementary summary. Yet the multidisciplinary nature of participatory studies, the inherent data variability, and their small sample sizes render them complex in design, analysis, and interpretation. Using three simple sets of ranking data from PRA studies, it is shown that statistical methods must be simple to be adopted, that data are often poor in quality, and that parametric assumptions might not always hold. The complex underlying structure of many datasets demands a flexible approach to analysis and interpretation. Greater collaboration between researchers and statisticians will improve the supply of good datasets that can be used to improve statistical techniques for this type of research and ultimately generate greater confidence in the use of statistical methodologies.