Precisão experimental em ensaios com a cultura do feijão

Abstract

The evaluation of the coefficient of variation (CV) as a precision measure of experiments has been carried out with several crops, animal species and forages through researches proposing classifying limits of the values, considering the mean, the standard deviation and the distribution of the CV values of various study variables involved in the experiments. The objective of this research was to study the distribution of the CV values of experiments with the bean crop, proposing limits that can guide the researchers in the evaluation of their studies with each variable. The data used were obtained from a review of scientific articles related to the bean crop. Several variables were considered: plant productivity, number of pods, number of grains per pod, weight of one-hundred grain sample, final stand, plant height, and harvest efficiency. Based on the normal distribution, limits of values of CV were obtained, also using the distribution of sample quantiles and the median and pseudosigma, classifying them as low, medium, high and very high. The statistical calculus was implemented using one of the functions of the free software R to verify the normality of the data. Results indicated that de CV values were different among all variables presenting huge variation, justifying the use of specific levels for each variable.