Abstract 010
Advanced Methods of Sensory Evaluation
European Brewery Convention Monograph VII, Flavor Symposium,
Copenhagen 1981 pp. 33-42
M.C. Meilgaard and K.J. Siebert
The applications of a number of statistical and multivariate methods to sensory evaluation are discussed. These include determining the degree of difference between several samples (Scheffe Paired Comparison Test, Multiple Paired comparison test, Signal Detection test); determining the intensity of perception in the suprathreshold range (Category Scaling, Magnitude Estimation); eliminating variation in sensory data between panelists (Procrustes Analysis); fitting equations to sensory and chemical data (Multiple Linear Regression, Stepwise Regression); grouping samples according to similarity or minimizing redundancy in terms (Cluster Analysis, Principal Components Analysis); devising equations that will separate samples into groups, ranking variables according to efficiency as discriminators, or detecting relationships in large arrays of data (Discriminant Analysis, Canonical Analysis, Factor Analysis); generating hypotheses on causal relationships between compounds and flavors (Multidimensional Scaling, Pattern Recognition).
Back to K.J. Siebert home page | Back to publication list