The sorting task: Stepwise
Before discussing the steps involved in the sorting task, it is important to keep the two simple questions that this procedure is trying to answer, in mind:
- Which products are similar and which differ in their sensory characteristics?
- Which sensory characteristics are responsible for the observed differences and similarities between the wines?
The answers to these two questions are elicited during the sorting task, in the following practical steps:
Step 1: Sorting of the wines in the sample set (which products are similar and which differ in their sensory characteristics?)
At the start of the sorting task, tasters receive all the wines simultaneously and are asked to group them according to similarity. Wines with similar sensory characteristics must be placed in the same group and wines with different characteristics, in different groups. Tasters must form more than one group. They may not put all the wines in a single group and each wine may not be placed in its own separate group.
There are several variations of the method. Two of these commonly used and providing good results for wine evaluation, are firstly free or uninformed sorting and secondly, informed or directed sorting. During a free sorting task, tasters can create any number of groups and use any of their own criteria to make groups. Informed or directed sorting applies when instructions are provided to the tasters on how to group wines. Typically, information pertaining to the number of groups in which the wines must be sorted, or the criteria to use for sorting the wines, can be provided. Colour, aroma, taste and mouth-feel can be assessed simultaneously or separately during different tasting sessions. It is often asked of tasters to use only aroma or taste during a sorting task.
A sorting task can be concluded at this stage, but often it is important to know which sensory characteristics are responsible for the perceived differences and similarities amongst the wines in the test set.
Step 2: Description of the main sensory characteristics of each group (which sensory characteristics are responsible for the differences and similarities?)
During the second part of the sorting task, tasters are asked to describe the sensory characteristics of the groups they formed, a procedure that is referred to as labelled sorting. It can be a very time consuming process to do statistical analysis on the data when tasters all use their own terms. A pre-determined list of sensory descriptions can be provided to the tasters and they can select the terms used to describe the groups best. This will simplify the task for the tasters, as well as ensure that they all use the same terms for a particular characteristic. In this way, the statistical analysis is kept relatively straightforward. Aroma wheels and sensory descriptions of similar wines from previous studies are examples of such lists.
Step 3: Statistical analysis and graphic projection of results
There are different approaches that can be used to analyse sorting data. The most common approach is to calculate (by counting) how many times each two wines were grouped together by all the tasters. The statistical method Multidimensional scaling (MDS) is used to produce a graph where wines that were grouped together by many tasters, are displayed close to each other, and wines often grouped into separate groups by the tasters, appear further apart.
The panel leader would typically reduce the number of descriptions given by the tasters, by combining terms with the similar meaning. Terms used by only one or two tasters are left out. The number of times a particular term was used by all the tasters to describe a wine, is then counted. Correspondence analysis (CA), are typically used to analyse data related to the descriptions. MDS and CA techniques can easily be done in XLSTAT software. There is expertise at IWBT-DVO, SU to support and train people who are interested to do these analyses.
Setting up a panel for the sorting task
One of the advantages of the sorting task is that training of a panel is not required. Wine experts, as well as people with lesser wine knowledge can participate as tasters. Numerous studies have been done with non-expert consumers. It is important to choose the right panel, in order to get the information required about the wines. Experts will more often be used if in-depth sensory profiles of the products are required. It may also be useful to do the same sorting experiment with both experts and non-expert consumers. An investigation into the similarities, and differences between the criteria used by the two groups to sort the wines, can be valuable for product and brand development.
Advantages and disadvantages of the Sorting method
Sorting is one of the best methods for the sensory characterisation of products such as wine and beer, when the differences between the products are clear, and the smaller differences are not so important. Research has shown that the results obtained with classical sensory Descriptive analysis (DA) and Sorting, often lead to the same conclusions. In some cases it will be necessary to rather do DA on the test set, if it is important to identify the less prominent sensory qualities in wines and quantify their intensities. The results Sorting produce are usually repeatable, even with non-expert users as tasters. The data analysis step can however be time consuming and it is sometimes difficult to interpret the descriptions of the groups formed by the tasters, if a pre-determined list was not used. The number of wines that can be tasted is limited, because the tasters receive all the wines to be tested simultaneously.
The Sorting method applied to wine: Some published examples
Free sorting of wines was used by Bester and co-authors in 2011 to show that wine experts and non-experts alike, find it difficult to distinguish between some South African Chenin blanc wine styles. This was particularly evident for the “fresh and fruity” and “rich and ripe” unwooded Chenin product categories. Mineral-like characteristics in wine, as well as the meaning that experts associate with the term, was researched by Ballester and co-authors in 2013 by using a sorting task. This research group also investigated the sorting of wines based on aroma in 2009. Sorting is becoming increasingly important in wine sensory evaluation, due to the fact that training of the panel is not required, and both experts, as well as ordinary consumers can be used as tasters.
The other articles in this series on wine sensory evaluation methods include Part 3 that deals with Projective mapping (PM) and Napping®, while Part 4 focusses on Check-all-that-apply (CATA). A comparison of the practical aspects and requirements of all the methods discussed in this series is also given in Part 4.
This research was made possible by Winetech funding, project IWBT W13/02: “Rapid descriptive sensory methods for wine evaluation – special focus on further optimisation of rapid methods and streamlining of workflow”. Winetech, the Institute for Grape and Wine Sciences (IGWS), Stellenbosch University, NRF and THRIP are acknowledged for financial support.
Consultation services offered by the Wine Sensory Facility, IWBT-DVO, Stellenbosch University:
- Selection of appropriate sensory methods.
- Guidance with selection, training, testing of panellists.
- Experimental design and practical workflow.
- Statistical analysis and interpretation of data.
– For more information, contact Jeanne Brand at firstname.lastname@example.org, Hélѐne Nieuwoudt at email@example.com or contact the authors at firstname.lastname@example.org.
Ballester, J., Abdi, H., Langlois, J., Peyron, D. & Valentin, D., 2009. The odor of colors: Can wine experts and novices distinguish the odors of white, red, and rosé wines? Chemosensory Perception 2, 203 – 213.
Ballester, J., Mihnea, M., Peyron, D. & Valentin, D., 2013. Exploring minerality of Burgundy Chardonnay wines: A sensory approach with wine experts and trained panellists. Australian Journal of Grape & Wine Research 19, 140 – 152.
Bester, I., 2011. Classifying South African Chenin blanc wine styles. Masters thesis. Institute for Wine Biotechnology, Department of Viticulture and Oenology, Stellenbosch University, South Africa.
Cartier, R., Rytz, A. & Lecomte, A., 2006. Sorting procedure as an alternative to quantitative descriptive analysis to obtain a product sensory map. Food Quality & Preference 17, 562 – 571.
Hulin, W.S. & Katz, D., 1935. The Frois-Wittmann pictures of facial expression. Journal of Experimental Psychology 18, 482 – 498.
Lawless, H.T., Sheng, N. & Knoops, S.S.C.P., 1995. Multidimensional scaling of sorting data applied to cheese perception. Food Quality and Preference 6, 91 – 98.
Lawless, H.T. & Heymann, H., 2010. Sensory evaluation of food: Principles and practices. New York: Springer.
Parr, W.V., Valentin, D., Green, J.A. & Dacremont, C., 2010. Evaluation of French and New Zealand Sauvignon wines by experienced French wine assessors. Food Quality & Preference 21, 56 – 64.
Valentin, D., Chollet, S., Lelievre, M. & Abdi, H., 2012. Quick and dirty, but still pretty good: A review of new descriptive methods in food science. International Journal of Food Science & Technology 47, 1563 – 1578.
Varela, P. & Ares, G., 2012. Sensory profiling, the blurred line between sensory and consumer science: A review of novel methods for product characterization. Food Research International 48, 893 – 908.