Tamanho de amostra de caracteres de Feijão Caupi

Abstract

The sample size represents the number of plants to be evaluated for appropriate experimental planning. It is assumed that there is variability of sample size between characters between varieties. Thus the purpose of this study was to determine sample size, for characters of caupi beans (Vigna unguiculata L. Walp.) of two cultivars and to check if variability in sample size. The blank test (experiment without treatment) was conducted. The experiment was conducted in the experimental area of the UFT - Campus Gurupi, with a total area of 360 m ² divided into two areas of 180 m ² using the cultivars BRS Guariba and BRS Nova Era. Each experimental unit consisted of 24 lines 15 m long, with a spacing of 0.50 m between rows. Six characters were evaluated (plant height, number of side branches, number of vargem per plant, length of vargem per plant, average number of vargem per plant, weight of grain per plant). For each character, descriptive statistics were calculated by observing the values of minimum, maximum, mean, standard error, standard deviation, and coefficient of variation. The interactive process was performed with 2000 sampling, with replacement using different sample sizes (n) obtaining the minimum values, percentile 2.5%, mean, percentile 97.5% and maximum for each graphically plotted character. The sample size was determined by the number of plants from which the confidence interval amplitude of 95% was less than or equal to the limit of 10 (higher accuracy), 15, 20, 25, 30, 35% (lower accuracy) of the mean estimate. It was possible to conclude that it hears variability between NRL and CVP characters. The characters of NVP, NGP and PGP do not have variability between characters and varieties where it is only possible to notice difference in variability in sample accuracy of 20% and 25% respectively.