Multivariate Techniques for Sensory and Consumer Studies

Training Provided by Creascience This 3-day session is targeted towards sensory and consumer scientists interested in an applied workshop on multivariate data analysis of sensory and consumer test data with a strong emphasis on graphical summary representations. During the first two days, classical multivariate techniques including Principal Component Analysis (PCA), Factor Analysis, Correspondence Analysis, Cluster Analysis and Discriminant Analysis are presented and their usefulness for sensory and consumer data is illustrated with case studies based on real data. The last day is dedicated to alternative and less known applications of these methods to address specific issues and get more actionable results from the collected data. For instance: PCA to measure panelist agreement in sensory attributes understanding Extended Preference Mapping for the identification of niche products & sensory drivers Discriminant analysis for product and concept mapping A combination of PCA and cluster analysis to select a subset of representative products in a larger series of samples. Throughout the workshop, participants are encouraged to use their own software and data to apply the multivariate techniques. Experienced instructors who have more than 15 years experience in the design and analysis of sensory and consumer test data will assist participants. They are very knowledgeable and proficient with most statistical packages.

Multivariate Techniques for Sensory and Consumer Studies

Day 1

• Principal Component Analysis (PCA): A Tool to Identify and Depict Data Redundancy
• Common Misinterpretations of PCA Biplots
• Applications of PCA for visual panel performance assessment
• Applications of PCA for consumer mapping with Internal Preference Mapping
• Factor Analysis (FA): A Tool to Extract Latent Variables/Underlying Dimensions
• The similarities and dissimilarities of PCA and Factor Analysis. When should each technique be used, and why?
• Applications of FA for determining the data dimensionality
• Applications of FA for extracting products' underlying dimensions
• Simple and Multiple Correspondence Analysis (SCA and MCA)
• The correct interpretation of SCA and MCA Biplots
• Applications of SCA and MCA for quantitative marketing projects and for the visualisation of questionnaire data

Day 2

• Cluster Analysis: A Tool to Create Groups of Homogeneous Objects
• Agglomerative Hierarchical Clustering Methods and K-Means
• Applications of cluster analysis for customer segmentation, product mapping and creating groups of products
• Discriminant Analysis (DA): A Tool to Identify the Most Important Characteristics that Help Distinguish Products/Concepts
• Applications of DA for finding the most discriminating product characteristics and mapping of products and concepts

Day 3

• Original Methods and Applications of Multivariate Techniques in Sensory and Consumer Research
• Applications of PCA for the idenfication of sensory drivers with Extended Preference Mapping
• Applications of CA for the idenfication of the ideal product/concept characteristics
• Applications of PCA and Cluster analysis for customer segmentation visualization
• Hands-On Applications: Sample datasets will be given to participants. However, participants are strongly encouraged to bring their own laptop, favourite statistical package and data to fully harness the power of multivariate data analysis tools.

About The Training Provider: Creascience