Mapping Hedonic Data: A Process Perspective

Abstract

Multivariate analyses are commonly used to study differences among items in a multidimensional space and to relate these findings to hedonic assessments of the same items. But there are numerous methods in use and the purpose of this article is to review these methods from a process standpoint. Specifically, this article considers the process assumptions behind several of the popular methods for multivariate mapping of hedonic data and argues that experimenters should consider how their data arise so that they can correctly interpret their findings. Among the methods considered in this article are models based on the hedonic continuum, internal and external preference mapping, and deterministic and probabilistic unfolding of preference and liking. Practical Applications Multivariate mapping of hedonic data has led to improved consumer products and a better understanding of consumer liking and choice. In this article, practitioners will find guidance in their choice of methods through a consideration of the processes that generate their data. Without a process‐based perspective, practitioners will not be able to optimally interpret results from the wide variety of available multivariate mapping methods.

Preference mapping of dulce de leche commercialized in Brazilian markets

“…Logistic regression (LR) is a powerful tool for the statistical analysis of categorical data with the main purpose to discover exploratory variables that drive a choice outcome, without necessity of evaluating integrals (Ennis and Ennis, 2013). It can be used to cor- Adhesiveness Ability of the product to adhere to the spoon or spatula when lying on the surface of the product, using only its own weight, and raised manually up to 90°L ow = dulce de leche ( (Prinyawiwatkul and Chompreeda, 2007).…”

“…Multivariate statistical techniques are very useful for analyzing complex data obtained from consumer and trained panelists, as they are able to provide a spatial representation of the experimental data (Ennis and Ennis, 2013;Zielinski et al, 2014). In addition, they proportionate the acquisition of the drivers of consumer liking and purchase intent, which is a concept frequently used in sensory and consumer research (Bi, 2012).…”

Prebiotic Low Sugar Chocolate Dairy Desserts: Physical and Optical Characteristics and Performance of PARAFAC and PCA Preference Map

“…This occurs as consumers sometimes do not translate adequately their samples’ perceptions into an overall score; 2nd, they may not find any sensory difference among the samples tasted and, accordingly, identical scores are assigned; and finally, the sources of variation in some consumers´ scores may need more than 2 first principal components to be captured (Resano and others ). Regards IMP, it is interesting emphasized that individuals do not have absolute ideal points because they may vary from time to time depending on variables such as mood, time of day, and recent consumption experiences in addition with the perception of these products by people at different moments (Enis and Enis ). In this sense, it would be preferable to derive the drivers of liking space directly from the hedonic data, if possible.…”

“…In this sense, it would be preferable to derive the drivers of liking space directly from the hedonic data, if possible. Such an approach has the advantage of implicitly using only those variables that have driven the hedonic responses (Enis and Enis ).…”

Applying parallel factor analysis and Tucker‐3 methods on sensory and instrumental data to establish preference maps: case study on sweet corn varieties

“…The term ‘unfolding’ comes from the idea of creating maps based on unidimensional data, which can be probabilities, ratings or ranks. These maps convey information about ideal and product locations …”