Introduction to component-based methods in sensory evaluation
This tutorial surveys some of the most frequently used methods for exploring multivariate data sets from
sensory and consumer science. Component-based methods are often applied for data reduction and
visualization of results.
The tutorial is divided into three parts.
In Part 1, we contrast principal component analysis (PCA) with principal variable analysis, then discuss
multiple factor analysis (MFA), which is often used to investigate multiblock sensory evaluation data of
various formats.
Part 2 focuses on the application of multivariate regression methods for sensory-instrumental correlations,
discussing principal component regression (PCR) and partial least squares regression (PLSR).
In Part 3, we discuss applications of ANOVA-simultaneous component analysis (ASCA) for the exploration of sensory data sets and temporal sensory data sets. ASCA involves partitioning the variance into main and interacting factors in an ANOVA-like manner and decomposing each variance partition into components.
Methods will be illustrated by example using free software such as R.
Castura, J.C., & Ricci, M. (2024). Introduction to component-based methods in sensory evaluation. 11th European Conference on Sensory and Consumer Research (Sensometrics Tutorial Day), September 8. Dublin, Ireland. (Tutorial).