Premium
RELATING CONSUMER ACCEPTANCE TO DESCRIPTIVE ATTRIBUTES BY THREE‐WAY EXTERNAL PREFERENCE MAPPING OBTAINED BY PARALLEL FACTOR ANALYSIS (PARAFAC)
Author(s) -
NUNES CLEITON A.,
BASTOS SABRINA C.,
PINHEIRO ANA CARLA M.,
PIMENTA CARLOS J.,
PIMENTA MARIA EMÍLIA S. GOMES
Publication year - 2012
Publication title -
journal of sensory studies
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.61
H-Index - 53
eISSN - 1745-459X
pISSN - 0887-8250
DOI - 10.1111/j.1745-459x.2012.00387.x
Subject(s) - descriptive statistics , preference , set (abstract data type) , acceptance testing , computer science , test (biology) , psychology , statistics , mathematics , paleontology , software engineering , biology , programming language
This work considers the hypothesis of using Parallel Factor Analysis (PARAFAC) for relating consumer acceptance to descriptive attributes by employing three‐way external preference maps. A pilot case study with a real data set from six commercial samples of grape juice was conducted. The results revealed that PARAFAC was useful for relating consumer acceptance to descriptive attributes, by simultaneously considering the various analyzed attributes by three‐way external preference maps. The three‐way external preference maps provided a general interpretation of the data, thereby making it possible to simultaneously relate consumer preferences, sensorial attributes and descriptive characteristics. PRACTICAL APPLICATIONS Three‐way external preference mapping enables the simultaneous analysis of the correlations (interactions) among consumer preferences, products and different evaluated attributes, as in the previously proposed three‐way internal preference mapping; but in this approach, descriptive parameters are also considered in the analysis. It enables an analysis of the overall performance of the samples in the acceptance test using a single map and to simultaneously correlate the descriptive attributes to acceptance for all consumer attributes. The three‐way approach is able to access overall performance of samples in the consumer acceptance test without the need to evaluate overall liking attribute. This is advantageous, e.g., to prevent the halo effect of attributes on overall liking.