Eficiência experimental em ensaios de VCU de feijão comum no Estado de Pernambuco

Abstract

In Brazil, several genotype competition experiments are conducted every year aiming at the recommendation of improved cultivars. In these tests, the requirements established by the Ministry of Agriculture, Livestock and Supply must be followed, which regulates these experiments and denominates as Value of Cultivation and Use (VCU). The common bean (Phaseolus vulgaris L.) VCU standards recommend that experiments should be conducted in at least three locations in the representative crops of each region for two years. Recent studies have pointed to the inadequate use of the coefficient of variation in the evaluation of experiments as the sole criterion of experimental precision. With the absence of studies in the experimental area of VCU for the State of Pernambuco, the objective of this work was to evaluate strategies for the evaluation of the experimental precision of trials of common bean lines in the state of Pernambuco, seeking to improve the use of tools CV, F and Selective Accuracy, for the selection of new cultivars. Recent work suggests that in order to achieve an accuracy target of 90%, the F test values associated with the effects of cultivars in the analysis of variance should be greater than 5.0. Based on the results obtained it can be affirmed that 60.31% of the trials of value of cultivation and use of common beans have high and very high precision. This information may in the future be used when the VCU is revised by the Ministry of Agriculture, Livestock and Supply to guide breeders.