The economy of human and financial resources, without losing the experimental precision, is considered an important factor in the planning of experiments. The objective of this research was to determine how many evaluations (replications) were necessary to predict the performance of common bean (Phaseolus vulgaris) cultivars. Grains yield of fourteen common beans cultivars were evaluated in nine experiments conducted at Santa Maria, in Rio Grande do Sul State, Brazil (latitude 29°42'S, longitude 53°49'W, altitude 95m) in the agricultural years of 2000/2001 and 2004/2005. The repeatability estimations were obtained through variance, structural and principal component analysis of number repetitions and it was calculated. Assays with four replications are sufficient to identify superior cultivars of the common beans, in relation to the grains yield, with 85% real accuracy.
Phaseolus vulgaris L.; repeatability; selection; experimental precision