Dimensionamento amostral e determinação do tamanho ótimo de parcelas para avaliação de mudas de berinjela e jiló

• Ano de defesa: 2017
• Autor(a) principal: Hell, Leonardo Raasch
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Espírito Santo; BR; Mestrado em Agricultura Tropical; UFES; Programa de Pós-Graduação em Agricultura Tropical
• Palavras-chave em Português: Solanum melongena; Solanum gilo; Unidade experimental; Amostragem; Precisão experimental; Plot; Sampling; Experimental accuracy; Produtividade agrícola; Planejamento experimental; Jiló; Berinjela; Agronomia; 63
• Link de acesso: http://repositorio.ufes.br/handle/10/8220

The number of plants evaluated in experiments should be a representative sample of the population. Their data must be reliable so that they allow high likelihood of success in subsequent experiments. Add to this the fact that the specialized literature does not yet provide answers on plot and sample size for most crops in the seedling phase. The aim of this study was to determine the size of plots and of the sample in the evaluation of eggplant and jilo seedlings. For this, two experiments were carried out in the municipality of Colatina - ES, where both vegetable seedlings were produced in trays of expanded polystyrene containing 128 cells. The first objective was to determine the sample size needed to estimate the average quality characteristics of eggplant and jilo seedlings applied to a set of characteristics (leaf number, total leaf area, shoot fresh matter mass, root fresh matter mass, total fresh matter mass, and Dickson quality index). Central trend and variability measures were calculated, and normality of the sample data was verified and then the sample size was calculated by bootstrap simulation. The sampling design requirement is different among the different characteristics of the eggplant and jilo seedlings, and also different for the same characteristic between the two species. The sample size for evaluating seedlings, for the estimation error of 10% of the estimated mean, with a x confidence level of 95%, is 32 and 26 eggplant and jilo seedlings, respectively. The second one had the objective of determining the optimum plot size for experiments involving eggplant and eggplant Solanaceae, applied on a set of characteristics (shoot height, stem diameter, number of leaves, total leaf area, shoot dry matter mass, root dry matter mass, total dry matter mass, and Dickson quality index). Measurements of central tendency and variability of the sample data were calculated and the optimal plot size was calculated using the modified maximum curvature method with bootstrap simulation. The optimal plot size is different for the characteristics evaluated in eggplant and jilo seedlings. Considering the evaluation of all aerial and root characteristics, the optimal plot size is 6 and 5 seedlings, for eggplant and jilo, respectively.