A common task for trained sensory assessors is to evaluate samples by characterizing and quantifying the intensities of sensory attributes. In some cases, the samples are instances of in-market or prototype products. To explore results from the panel, it is conventional to obtain a products-by-attributes table of means, center and variance-standardize its columns, then conduct principal component analysis (PCA). The principal components that extract variance maximally from the products also extract variance maximally from all product paired comparisons. However, if there is a successful in-market reference product and multiple prototypes, then only the reference-prototype paired comparisons are of primary interest. PCA conducted conventionally does not extract variance maximally from the subset of reference-prototype pairs. To investigate a subset of paired comparisons, we crossdiff-unfold the results matrix, then keep only the rows for paired comparisons that are of primary interest, which becomes the input matrix for PCA. Advantages of PCA of this modified input matrix are demonstrated visually and numerically using a case study involving cheddar cheeses. Data sets with related structures, including temporal sensory data sets and data sets in domains outside sensory evaluation can also be investigated using this approach.<\/p>\n\n\n\n
Castura, J.C., Varela, P. & N\u00e6s, T. (2023). Principal component analysis of sensory panel results for a reference and multiple prototypes. 2023 Joint Statistical Meetings (JSM).<\/em> 5-10 August. Toronto, Canada. (Poster Presentation). Download<\/a><\/p>\n","protected":false},"excerpt":{"rendered":" A common task for trained sensory assessors is to evaluate samples by characterizing and quantifying the intensities of sensory attributes. In some cases, the samples are instances of in-market or prototype products. To explore results from the panel, it is conventional to obtain a products-by-attributes table of means, center and variance-standardize its columns, then conduct principal component analysis (PCA). The<\/p>\n