Sensory analysis of Prato cheeses by generalized linear mixed models

• Autor(a) principal: Alvarenga, Tatiane Carvalho
• Data de Publicação: 2022
• Outros Autores: Rodrigues, Jéssica Ferreira
• Tipo de documento: Artigo
• Idioma: por
• Título da fonte: Revista do Instituto de Laticínios Cândido Tostes
• Texto Completo: https://www.revistadoilct.com.br/rilct/article/view/892
• topic: analysis of categorized data; proportional odds models; nested models.; análise de dados categorizados; modelos de chances proporcionais. modelos encaixados.

Sensory analysis, an area of Food Science, is used to analyze and measure characteristics of foods, being able to evaluate the acceptance of samples. Such assessments can be performed using the 9-point numerical hedonic scale, classified as an ordinal categorized variable. To verify the significance of each effect involved in the sensorial analysis, the experiment was carried out at the Department of Food Science at the Federal University of Lavras, with two brands of Prato cheese evaluated by 100 evaluators, each male and female. For this, fixed and random effect models were used, considering the effects of cheese and sex and the interaction between both fixed effects, and the effect of the different evaluators was considered as a random effect because they are the repetitions in the experiment. It was concluded that the effects of cheese, sex, and the interaction between them and the evaluator effect were all significant in the model (AIC=1465), showing that the best model for analyzing the grades regarding the acceptance of cheeses was the generalized mixed model. Brand A had better acceptance in terms of preference (average score on the 7-point hedonic scale).