Sensory profile of gorgonzola via generalized procrustes analysis using R

ABSTRACT

Sensory data analysis in Brazil used to be performed through univariate Analysis of Variance (ANOVAs) hence one looses valuable intra variable information. To solve this porblem one may use multivariate tools. Generalized Procrustes Analysis (GPA), often performed in expensive software, is a multivariate exploratory data analysis method that aims: i) remove scores bias; ii) resume the assessors agreement, usually, on a plant; iii) show the objects relative differences. An application on sensory profile of Gorgonzola cheese is porvided. Data was analyzed by uni, multivariate ANOVAs and GPA, on the free statistical software R, ANOVAs detected significant differences among cheeses and GPA provided the consensus configuration displaying their relative differences and sensory evolution. Assesors preferred cheeses with a higher fat content. GPA can be reasonably performed by free software, say R, and should be utilized like a complement of Analysis of Variance.