Dimensionamento amostral e tamanho ótimo de parcela na experimentação com mudas clonais de cacaueiro
Carregando...
Data
2025-09-26
Autores
Morais, Letícia Galvão
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Federal do Espírito Santo
Resumo
Cocoa is a crop of great economic importance for Brazil and has been economically exploited since the 1930s. Therefore, the crop has undergone and continues to undergo several improvement processes that have allowed its production to be diversified. For this to continue, scientific experiments are essential, especially those aimed at evaluating the quality of seedlings, since this is the main form of crop propagation, and there is a specific number of clonal varieties. In this type of experiment, there is no consensus regarding the amount of material necessary to obtain reliable results, moreover, when evaluating a characteristic, it is common to observe variation between plants, even among those subjected to the same treatment. Therefore, it is important to base on a scientific justification to determine the amount of material that will be used to carry out effective experimental planning. In this sense, the objective was to determine, through two different methodologies, the amount of material, that is, the number of plants, that should be used in the experimentation with clonal cocoa seedlings. Therefore, the sample size was determined using the deterministic method, and the optimal plot size was determined using the modified maximum curvature method with bootstrap simulation. In both cases, 50 seedlings from six different clones were used, totaling 300 plants. 12 morphological traits and four indexes used in seedling morphological characterization studies were evaluated. Descriptive statistics were performed, including: mean, median, maximum, minimum, standard deviation, and coefficient of variation. For the first chapter, the sample size was also determined using the Kolmogorov-Smirnov normality test and skewness. A sample size was determined for estimating errors set at 5%, 10%, 15%, 20%, 25%, and 30% of the mean, with a 95% confidence level. The sample size for evaluating traits obtained by measurement, weighing, leaf number, leaf area, root number, and volume increases in this order. For the indexes, the sample size increases in the order of IR, IRC, IQD, and AFE. Allowing for a 20% error, the sample size of 119 seedlings covers the six clones used in this study, as well as the sixteen variables evaluated. In the second chapter, the optimal plot size for experimentation with clonal cacao seedlings was determined, and the value found was nine plants per experimental plot.
Descrição
Palavras-chave
Theobroma cacao L. , Bootstrap , Precisão experimental