A chemometrics study of analytical panels in virgin olive oil. An approach for evaluating panels in training

R Aparicio,J Rodríguez,F Gutiérrez

doi:10.3989/gya.1991.v42.i3.1245

Abstract

A mathematical procedure including Canonical Correlation, Factor and cluster analysis and Multiple Regression was designed to study whether two different sensory panels, with different levels of training, could be considered as a unity. Among others, correlations between each attribute of both panels, matrices of partial correlations, variance explained by the most important factors and similarity among clusters were computed as the procedure to evaluate if a Panel in training works as well as an expertise. In addition, every one of the factors was correlated with the final evaluation of quality for analyzing the interrelationship between the profile notes and the grading table of virgin olive oil quality from a mathematical point of view. Finally, a stepwise multiple correlation analyzes how the panels modeled the evaluated samples. Promising directions for future research are also suggested.

Highlights

In 1974 Stoner et al published the quantitative descriptive analysis method (QDA) opening a new way in sensorial evaluation
It was not possible to analyse mathematically the amount of information gathered by QDA without statistical multivariate algorithms
Sensory and quality problems are generally multivariate problems (Resurrecion, 1988b), (Powers, 1988). Factor analysis procedures such as principal components or principal factor (Clapperton et al, 1979), stepwise discriminant analysis (Ennis et al, 1980), different kinds of cluster programs (Sheppard, 1980), (Resurrecion, 1988a) and multiple regression (Resurrección, 1988b) have been some of the statistical methods widely used in reaching conclusions

Summary

Introduction

In 1974 Stoner et al published the quantitative descriptive analysis method (QDA) opening a new way in sensorial evaluation. The consequence was an increasing use of the descriptive and quantitative multiscalar test in the sensorial analysis of foods that has become as widely used as Bengtsson's triangular test (Bengtsson, 1943). Sensory and quality problems are generally multivariate problems (Resurrecion, 1988b), (Powers, 1988). Factor analysis procedures such as principal components or principal factor (Clapperton et al, 1979), stepwise discriminant analysis (Ennis et al, 1980), different kinds of cluster programs (Sheppard, 1980), (Resurrecion, 1988a) and multiple regression (Resurrección, 1988b) have been some of the statistical methods widely used in reaching conclusions. The autors consider that the panel, constituted by human panelists, is the sensory measurement instrument and try evaluating it

Methods

Results

Conclusion