Statistics for Sensory and Consumer Science

This book will describe the most basic and used statistical methods for analysis of data from trained sensory panels and consumer panels with a focus on applications of the methods. It will start with a chapter discussing the differences and similarities between data from trained sensory and consumer tests. Different uses and the potential of the sensory/consumer methods will be stressed. Aspects related to choice of measurement scale and consequences for choice of statistical method will also be discussed in the same chapter. The main part of the book will be devoted to describing the most basic methods used in the sensory area and also how they are linked to various important applications. This will be organised by splitting the book in two, one part describing important problems met and studied within sensory and consumer science accompanied with a short description of the statistical methods needed and how to apply them. The second part will be more statistically oriented and will describe the various statistical methods in more detail. It will be possible to read the two parts separately, but one of the main benefits of the book is that the two are presented adjacent to each other and with substantial cross-referencing between them. In addition, the book will contain a section relating the methodology to established software packages and it will describe how the different methods can be applied.

Professor Tormod Naes is a Principal Research Scientist based at Matforsk, a government food research laboratory, in Norway. He received his PhD in statistics from University of Oslo in 1984. He is also currently employed as a Professor at the Institute of Mathematics at the University of Oslo. He serves on the editorial boards of Journal of Chemometrics, Journal of Near Infrared Spectroscopy and Food Quality and Preference.
His main area of research is the development and use of multivariate statistical methods in food science. In particular in applications within the areas of sensory analysis, spectroscopy, process optimisation and bioinformatics. He has published 108 refereed papers and co-authored and co-edited 5 books in multivariate analysis and analysis of variance, including the highly cited "Multivariate Calibration" co-authored with Professor Harald Martens (Wiley 1988). He has received the Tomas Hirschfeld award in NIR analysis. (1997), EAS award for achievements in chemometrics (1997), Kowalski award in Chemometrics (J. Wiley and Sons) (2006) and is an Honorary member of the Chemometric Society of Norway (2006).

Preface	p. ix
Acknowledgements	p. xi
Introduction	p. 1
The Distinction between Trained Sensory Panels and Consumer Panels	p. 1
The Need fox Statistics in Experimental Planning and Analysis	p. 2
Scales and Data Types	p. 3
Organisation of the Book	p. 3
Important Data Collection Techniques for Sensory and Consumer Studies	p. 5
Sensory Panel Methodologies	p. 5
Consumer Tests	p. 7
Problem Driven
Quality Control of Sensory Profile Data	p. 11
General Introduction	p. 11
Visual Inspection of Raw Data	p. 15
Mixed Model ANOVA for Assessing the Importance of the Sensory Attributes	p. 18
Overall Assessment of Assessor Differences Using All Variables Simultaneously	p. 19
Methods for Detecting Differences in Use of the Scale	p. 24
Comparing the Assessors' Ability to Detect Differences between the Products	p. 27
Relations between Individual Assessor Ratings and the Panel Average	p. 29
Individual Line Plots for Detailed Inspection of Assessors	p. 33
Miscellaneous Methods	p. 34
Correction Methods and Other Remedies for Improving Sensory Profile Data	p. 39
Introduction	p. 39
Correcting for Different Use of the Scale	p. 40
Computing Improved Panel Averages	p. 43
Pre-processing of Data for Three-Way Analysis	p. 45
Detecting and Studying Sensory Differences and Similarities between Products	p. 47
Introduction	p. 47
Analysing Sensory Profile Data: Univariate Case	p. 48
Analysing Sensory Profile Data; Multivariate Case	p. 59
Relating Sensory Data to Other Measurements	p. 67
Introduction	p. 67
Estimating Relations between Consensus Profiles and External Data	p. 68
Estimating Relations between Individual Sensory Profiles and External Data	p. 74
Discrimination and Similarity Testing	p. 79
Introduction	p. 79
Analysis of Data from Basic Sensory Discrimination Tests	p. 80
Examples of Basic Discrimination Testing	p. 81
Power Calculations in Discrimination Tests	p. 85
Thurstonian Modelling: What Is It Really?	p. 86
Similarity versus Difference Testing	p. 87
Replications: What to Do?	p. 89
Designed Experiments, Extended Analysis and Other Test Protocols	p. 93
Investigating Important Factors Influencing Food Acceptance and Choice	p. 95
Introduction	p. 95
Preliminary Analysis of Consumer Data Sets (Raw Data Overview)	p. 99
Experimental Designs for Rating Based Consumer Studies	p. 102
Analysis of Categorical Effect Variables	p. 106
Incorporating Additional Information about Consumers	p. 113
Modelling of Factors as Continuous Variables	p. 117
Reliability/Validity Testing for Rating Based Methods	p. 118
Rank Based Methodology	p. 119
Choice Based Conjoint Analysis	p. 120
Market Share Simulation	p. 123
Preference Mapping for Understanding Relations between Sensory Product Attributes and Consumer Acceptance	p. 127
Introduction	p. 128
External and Internal Preference Mapping	p. 129
Examples of Linear Preference Mapping	p. 136
Ideal Point Preference Mapping	p. 141
Selecting Samples for Preference Mapping	p. 146
Incorporating Additional Consumer Attributes	p. 147
Combining Preference Mapping with Additional Information-about the Samples	p. 149
Segmentation of Consumer Data	p. 155
Introduction	p. 155
Segmentation of Rating Data	p. 156
Relating Segments to Consumer Attributes	p. 163
Method Oriented
Basic Statistics	p. 165
Basic Concepts and Principles	p. 165
Histogram, Frequency and Probability	p. 166
Some Basic Properties of a Distribution (Mean, Variance and Standard Deviation)	p. 168
Hypothesis Testing and Confidence Intervals for the Mean ¿	p. 169
Statistical Process Control	p. 172
Relationships between Two or More Variables	p. 173
Simple Linear Regression	p. 175
Binomial Distribution and Tests	p. 177
Contingency Tables and Homogeneity Testing	p. 178
Design of Experiments for Sensory and Consumer Data	p. 181
Introduction	p. 181
Important Concepts and Distinctions	p. 182
Full Factorial Designs	p. 185
Fractional Factorial Designs: Screening Designs	p. 187
Randomised Blocks and Incomplete Block Designs	p. 188
Split-Plot and Nested Designs	p. 190
Power of Experiments	p. 191
ANOVA for Sensory and Consumer Data	p. 193
Introduction	p. 193
One-Way ANOVA	p. 194
Single Replicate Two-Way ANOVA	p. 196
Two-Way ANOVA with Randomised Replications	p. 198
Multi-Way ANOVA	p. 200
ANOVA for Fractional Factorial Designs	p. 201
Fixed and Random Effects in ANOVA: Mixed Models	p. 203
Nested and Split-Plot Models	p. 205
Post Hoc Testing	p. 206
Principal Component Analysis	p. 209
Interpretation of Complex Data Sets by PCA	p. 209
Data Structures for the PCA	p. 210
PCA: Description of the Method	p. 211
Projections and Linear Combinations	p. 213
The Scores and Loadings Plots	p. 214
Correlation Loadings Plot	p. 217
Standardisation	p. 219
Calculations and Missing Values	p. 220
Validation	p. 220
Outlier Diagnostics	p. 221
Tucker-1	p. 223
The Relation between PCA and Factor Analysis (FA)	p. 224
Multiple Regression, Principal Components Regression and Partial Least Squares Regression	p. 227
Introduction	p. 227
Multivariate Linear Regression	p. 229
The Relation between ANOVA and Regression Analysis	p. 232
Linear Regression Used for Estimating Polynomial Models	p. 233
Combining Continuous and Categorical Variables	p. 234
Variable Selection for Multiple Linear Regression	p. 235
Principal Components Regression (PCR)	p. 236
Partial Least Squares (PLS) Regression	p. 237
Model Validation: Prediction Performance	p. 238
Model Diagnostics and Outlier Detection	p. 241
Discriminant Analysis	p. 244
Generalised Linear Models- Logistic Regression and Multinomial Regression	p. 245
Cluster Analysis: Unsupervised Classification	p. 249
Introduction	p. 249
Hierarchical Clustering	p. 251
Partitioning Methods	p. 254
Cluster Analysis for Matrices	p. 259
Miscellaneous Methodologies	p. 263
Three-Way Analysis of Sensory Data	p. 263
Relating Three-Way Data to Two-Way Data	p. 269
Path Modelling	p. 269
MDS-Multidimensional Scaling	p. 271
Analysing Rank Data	p. 271
The L-PLS Method	p. 273
Missing Value Estimation	p. 273
Nomenclature, Symbols and Abbreviations	p. 277
Index	p. 283
Table of Contents provided by Ingram. All Rights Reserved.

What is included with this book?

The New copy of this book will include any supplemental materials advertised. Please check the title of the book to determine if it should include any access cards, study guides, lab manuals, CDs, etc.

The Used, Rental and eBook copies of this book are not guaranteed to include any supplemental materials. Typically, only the book itself is included. This is true even if the title states it includes any access cards, study guides, lab manuals, CDs, etc.

FREE SHIPPING

Amazon no longer offers textbook rentals. We do!

Statistics for Sensory and Consumer Science

Summary

Author Biography

Table of Contents

Supplemental Materials