Statistical tools for farmers, participatory trials for conservation agriculture
2009
Parsad, R. | Crossa, J. | Gupta, V.K. | Gupta, R.K. | Ladha, J.K. | Anitha, R.K.
The Rice-Wheat Consortium (RWC) for the Indo-Gangetic Plains generally evaluates different resource-conserving technologies (RCTs) in a farmers participatory mode. It is a key challenge to analyze unbalanced data (due to variations in the number of replications of different RCTs, the use of different varieties, farmers preferences for testing different options, etc.) generated from these trials using proper statistical tools so as to draw meaningful and valid conclusions. To achieve this, data collection and data preparation are of paramount importance. An additional problem with the analysis of data generated from these trials is that different researchers use different terms for similar practices. We reiterated the use of uniform terminology for RCTs and experimental variables and a common data entry format, including specified units. The usefulness of the linear mixed effects model has also been emphasized to analyze the data generated from these trials by taking farmer effects or field effects as random and RCT effects (henceforth called treatments) as fixed. To identify the best performance of any treatment in a given environment, one can make all possible pair-wise treatment comparisons using adjusted mean/best linear unbiased predictors of treatments. The procedure of analyzing groups of experiments can be used to study the interaction of treatments with crop varieties, years, soil types, and land leveling, etc. In the case of crossover interactions, the site regression biplot technique has been suggested to identify subsets of treatments to be recommended for specific environments. All the techniques developed/suggested have been illustrated with examples.
显示更多 [+] 显示较少 [-]