In Brazil, numerous field trials are conducted every year with the aim of recommending improved cultivars. These trials must comply with the requirements established by the Ministry of Agriculture, Livestock and Supply, the body that standardises these experiments and calls them Cultivation and Use Value (VCU).

The VCU rules for common beans (Phaseolus vulgaris L. ) stipulate that the experiments must be conducted in at least three locations, during the representative harvests of each region, for two years.

Recent studies have pointed out the problems of considering only the value of the experimental coefficient of variation when assessing the precision of an experiment. Given the lack of studies in the area of experimental precision for the state of Pernambuco, the aim of this study was to evaluate a new strategy for assessing the experimental precision of trials of common bean lines in the state of Pernambuco, with the aim of improving the use of the statistical tools CV, F and Selective Accuracy for selecting new cultivars.

. .