The appropriate sample size for the evaluation of morphological fruit traits of pepper was evaluated through a technique of simulation of subsamples. The treatments consisted of eight accessions of four pepper species (Capsicum spp.), cultivated in an experimental area of the Universidade Federal da Paraíba. Small samples, ranging from 3 to 29 fruits were evaluated. For each sample size, 100 subsamples were simulated with data replacement. The data were submitted to analysis of variance, in a complete randomized design, for the minimum number of fruits per sample. Each collected data consisted of the first number of fruits in the simulated sample without values out of the confidence interval. This procedure was done up to the last subsample simulation. The simulation technique provided reductions in the sample size, around 50%, with the same accuracy of the sample of 30 fruits and the accessions did not differ significantly.
Capsicum spp; subsamples; simulation
