Perfil descritivo otimizado: número de julgadores, delineamentos e validação

• Autor(a) principal: Silva, Rita de Cássia dos Santos Navarro da
• Data de Publicação: 2013
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: LOCUS Repositório Institucional da UFV
• Texto Completo: http://locus.ufv.br/handle/123456789/489
• topic: Análise descritiva; Perfil descritivo otimizado; Julgadores semi-treinados; Métodos rápidos; Validação; Descriptive analysis; Optimized descriptive profile; Semi-trained judges; Fast methods; Validation; CNPQ::CIENCIAS AGRARIAS::CIENCIA E TECNOLOGIA DE ALIMENTOS::CIENCIA DE ALIMENTOS

The Optimized Descriptive Profile (ODP) is a new methodology for sensory description that was recently proposed in order to reduce the time of sensory testing. In the ODP, the evaluation of sensory attributes is performed by means of quantitative evaluation by a semi-trained panel. Evaluation of attributed intensity on an interval scale permits identification of the magnitude of difference between products, increasing applicability of the method as well as its relevance among fast sensory techniques. The purpose of this study was to validate the ODP technique as a descriptive methodology, as well as determine the optimal experimental conditions for application of the method, including studies on determining the number of judges and comparison of experimental designs. In determining the ideal number of judges, a computer simulation study was conducted, using the re-sampling technique for data from an original evaluating panel, considering 10,000 sub-groups with replacement. The criteria considered were: (i) obtaining an experimental error less than or equal to the error found in the reference method (Conventional Profile), (ii) discrimination of samples as discriminated obtained by the complete panel, and (iii) minimal loss of information in the sensory map. For evaluation of criterion (i) a literature review was performed to collect experimental error values obtained in previous studies with the Conventional Profile, establishing a cut-off for determining the optimal number of judges in the ODP. In analyzing the results, it was found that at least sixteen judges should make up the sensory panel so that the ODP meets the criteria for magnitude of the experimental error, where the remaining criteria were fully met when this number of judges was used. In the study of experimental designs, was evaluated the feasibility of using a Balanced Incomplete Block Design (BIBD) to collect data according to the protocol of the ODP. Thus, an identical set of samples was evaluated by three independent teams, where each team evaluated the samples using a specific methodology and design. Two panels evaluated the samples using the ODP technique, where one panel used the Balanced Complete Block Design (BCBD) and the other the BIBD. A third panel evaluated the samples according to the Conventional Profile technique (CP) using a trained judges panel. The mean scores obtained in the different techniques/designs showed no significant difference by the F- test (p > 0.05), considering the different sensory attributes. A high degree of similarity was found between the sensory maps obtained by different techniques/designs, presenting RV coefficients greater than 0.95. The DBIB is presented as an alternative to collecting data in the ODP without loss in quality of the results. In the validation study of the ODP methodology, accuracy, precision (repeatability and reproducibility) and robustness were measured. Accuracy of the data generated by the ODP was measured by comparison with the reference method (CP), considering different food matrices. Precision was measured at the panel (repeatability) and inter-laboratory levels (reproducibility). Robustness considered evaluation of the sensitivity of the method with regards to reducing the number of judges making up the panel. In measuring of the validation criteria, the ODP methodology met the established criteria, presenting a degree of proximity with the reference methodology greater than 95%. In evaluation of measurement repeatability, considering three repetitions of the evaluations by a single panel, the ODP showed statistical equality among the repetitions (p > 0.05) with degree of proximity exceeding 99%. This method also showed reproducibility of the data when laboratory were altered, presenting a perfect combination among sensory profiles. The ODP showed to be robust in reducing the number of judges, where smaller teams presented less random variation than the complete panel. It was also successfully validated and presented validation measures with a high degree of certainty, considering different food matrices.