Multivariate Data Analysis in Sensory and Consumer Science

ISBN-13: 9780917678417 / Angielski / Twarda / 2004 / 320 str.

Garmt B. Dijksterhuis

Opis

This book is an outgrowth of research done by Dr. Gamt Dijsterhuis for his doctoral thesis at the University of Leiden. However, there are also contributions by several other authors, as well, including Eeke van der Burg, John Gower, Pieter Punter, Els van den Broek, and Margo Flipsen.
This book discusses the use of Multivariate Data Analysis to solve problems in sensory and consumer research. More specifically the focus is on the analysis of the reactions to certain characteristics of food products, which are in the form of scores given to attributes perceived in the food stimuli; the analyses are multivariate; and the senses are mainly the senses of smell and taste.
The four main themes covered in the book are: (1) Individual Differences, (2) Measurement Levels; (3) Sensory-Instrumental Relations, and (4) Time-Intensity Data Analysis.
The statistical methods discussed include Principle Components Analysis, Generalized Procrustes Analysis, Multidimensional Scaling, Redundancy Analysis, and Canonical Analysis.
This book will be a value to all professionals and students working in the sensory studies

Spis treści

Prologue and Acknowledgements.

Introduction..

Part I: Individual Differences.

Assessing Panel Consonance.

Interpreting Generalized Procrustes Analysis "Analysis of Variance" Tables..

Part II: Measurement Levels.

Multivariate Analysis of Coffee Images.

Nonlinear Canonical Correlation Analysis of Multiway Data.

Nonlinear Generalised Canonical Analysis: Introduction and Application from Sensory Research..

Part III: Sensory–Instrumental Relations.

An Application of Nonlinear Redundancy Analysis.

An Application of Nonlinear Redundancy Analysis and Canonical Correlation Analysis.

Procurstes Analysis in Studying Sensory–Instrumental Relations..

Part V: Time–Intensity Data Analysis.

Principal Component Analysis of Time–Intensity Bitterness Curves.

Principal Component Analysis of Time–Intensity Curves: Three Methods Compared.

Matching the Shape of Time–Intensity Curves.

Concluding Remarks.

References.

Abbreviations and Acronyms.

Author Index.

Subject Index.

O autorze

Edited by Garmt B. Dijksterhuis, Ph.D., ID–DLO, Institute for Animal Science and Health, Food Science Department, Lelystad, The Netherlands

Blurb

This book is about Sensometrics: the statistical analysis of data from sensory and consumer science. Sensory and consumer science is a broad field, comprising research in food, drinks, personal care products, tobacco, cleaning agents, etc. Before a product is to be launched on the market it is often tested by sensory panels. The analytical panels with trained judges assess the products on a great number of specific properties using sensory profiling techniques. In consumer tests the focus is on the acceptance of certain products or the preference for a specific product.

Following an introduction to sensory and consumer science and the special types of data abundant in these fields, eleven chapters address four themes in sensometrics: individual differences between the assessors in a sensory panel, the measurement level of the collected data and non linear analyses, sensory–instrumental relations, and time–intensity data analysis. The analyses are performed using, among others, Procrustes Analysis, Multidimensional Scaling, Principal Component Analysis, Canonical Analysis, Redundancy Analysis, and Biplots.