Free sorting

Actually, when considering the time needed for subjects to complete the evaluation task, the cards are being reshuffled. Although the simplest comparative tasks, as in Free Sorting, may still be faster, monadic sequential evaluation of the samples in... 

Successful attempts to collect product descriptions from professional sensory experts have relied on Flash Profile approaches (Eladan et al. (2005) with perfumers, Lassoued et al. (2008) with bakers and milling professionals, Dairou et al. (2003) with car pilots), on Napping and Projective Mapping (Perrin et al. (2008) with wine professionals, Nestrud and Lawless (2008) with culinary professionals), or on Free Sorting (Souffiet et al. (2004) with textile experts, Ballester et al. (2008) with wine experts). Several other examples are presented throughout this book. 

In this survey we focused on six methods conventional sensory profiling. Free Choice Profiling, Hash Profiling, CATA, Repertory Grid and Free Sorting. The questionnaire was built in three dimensions in order to assess, for each method, the level of use and of knowledge, the expected output when a method is used (in terms of quality, innovation, etc.), and the main strengths and weaknesses that were perceived in each case. 

Free Sorting Low level of use (34%) Good level of knowledge (65%) Well adapted to understand a competitor view -h task easy to understand, method easy to run and organize, can be achieved by different types of subjects (from experts to consumers), based on theory in psychology no dedicated software, data analysis is often an issue, needs a specific number of products (between 6 and 12), not adapted to monitor product quality... 

Besides, the sensory descriptions provided by these methods are usually not as accurate as those obtained by conventional profiling, though this criterion may not always apply (e.g. in the case of Free Sorting with consumers). This lower accuracy is notably linked to the more difficult interpretation when using free vocabulary. A direct consequence of this is the need for more advanced data analysis techniques. Eventually, the translation of sensory properties into technical product variables may be more difficult. 

Faye, P., Bremaud, D., Durand Daubin, M., Courcoux, P., Giboreau, A. and Nicod, H. (2004). Perceptive free sorting and verbalization tasks with naive subjects an alternative to descriptive mappings. Food Quality and Preference, 15, 781-791. 

Free sorting as a sensory profiling technique for product development... 

Depending on the objective of the study, free sorting may be performed considering one or several sensory modalities and the criterion for grouping the objects may be specified (texture, odor, etc.). In contrast, if the global similarity between stimuli is of interest, the subjects are advised that they are free to use any characteristic of the products to form coherent and homogeneous groups. 

Several variants of the free sorting task have been developed for specific objectives. Since they differ in terms of the instructions given to the subjects and the type of collected data, they can be considered as separate procedures. 

The methodology of multiple free sorting is detailed in Chapter 8 of this book. 

The holistic nature of free sorting confers on this task several advantages over more analytical procednres. Indeed, there is no need to generate an exhaustive list of... 

The multiple sorting task and the taxonomic sorting task are variants that allow the assessment of a more refined evaluation of the dissimilarities between stimuli at the subject level than the (single) free sorting task. 

As mentioned above, a free sorting test yields a family of partitions of the same set of products, each partition being associated with a particular subject. For the statistical treatment of free sorting data, several strategies of analysis have been proposed. We refer to the book by Coxon (1999) for a comprehensive review of these strategies. We also refer to a French paper by Faye et al. (2011) for a comparison of several methods of analysis on the basis of a case study. 

As an illustration, let us mention that the use of additive trees for analyzing free sorting data is relatively popular in psychology (Dubois, 1991), whereas in psychoacoustics, MDS methods are more often used (Gygi et al., 2007 MacAdams et al., 1995). In the field of sensory analysis that particularly interests us here, the mainstream is to use MDS methods (Faye et al., 2004 King et al., 1998 Lawless et al., 1995 Parr et al, 2007). However, alternative methods of analysis, such as multiple correspondences and allied methods, have also been proposed in this framework (Cadoret et al, 2009 Qannari et al., 2009 Takane, 1981,1982). 

With no claims to exhaustiveness, we present in the Table 7.1 some of the most important strategies of analysis that have been proposed to analyze free sorting data. 

Table 7.1 General overview of the statistical methods of analyzing free sorting data...

In free sorting task, let us consider the matrix A = (5y) of global dissimilarities between products. As stated above, the dissimilarity 5 between products i and j (i, j = 1,. .., n) is equal to the number of subjects who did not set these two products in the same group. Let us denote by X the representation space of dimension k (say). We suppose for the time being that the parameter k is fixed. We denote by di/X) the Euclidean distance between products i and j obtained from the coordinates of products i and j in the representation space, X. The optimal representation space is obtained by minimizing a criterion called Stress (Kruskal, 1964) ... 

We claim that non-metric MDS is better suited to analyzing free sorting data than metric MDS (Faye et al., 2004 Lawless et al., 1995). Our claim is backed up by the fact that the dissimilarities in matrix A are computed by counting the number of subjects who did not set the products in the same groups and, therefore, do not stem from interval or ratio scale data. 

We will also outline the representation of the products by means of an additive tree. Indeed, this is an appealing concept, which has been used in several studies related to the free sorting task. 

---