Reprinted, with permission, from ASTM E-253-Standard Terminology Relating to Sensory Evaluation of Materials and Products, copyright ASTM International, 100 Barr Harbor Drive, West Conshohocken, PA 19428. A copy of the ASTM standard may be obtained from ASTM International. http://www.astm.org/Standards/E253.htm
]]>Reprinted, with permission, from ASTM E-253-Standard Terminology Relating to Sensory Evaluation of Materials and Products, copyright ASTM International, 100 Barr Harbor Drive, West Conshohocken, PA 19428. A copy of the ASTM standard may be obtained from ASTM International. http://www.astm.org/Standards/E253.htm
]]>Reprinted, with permission, from ASTM E-253-Standard Terminology Relating to Sensory Evaluation of Materials and Products, copyright ASTM International, 100 Barr Harbor Drive, West Conshohocken, PA 19428. A copy of the ASTM standard may be obtained from ASTM International. http://www.astm.org/Standards/E253.htm
]]>Reprinted, with permission, from ASTM E-253-Standard Terminology Relating to Sensory Evaluation of Materials and Products, copyright ASTM International, 100 Barr Harbor Drive, West Conshohocken, PA 19428. A copy of the ASTM standard may be obtained from ASTM International. http://www.astm.org/Standards/E253.htm
]]>Please note that in order to create a new page or edit an existing wiki article, you must be a logged-in member of the society. Non-members are welcome to submit material to the wiki Editorial Board using the wiki submission form.
If are new to working with the wiki, you may want to read the New Article Tutorial. It gives the basic steps for creating a new wiki entry, including screen shots of making the Example Article piece. You should also read the article style piece as it explains several important formatting conventions. You can see the text behind any article by using the Edit menu tab (only visible when logged in), and is a good way to see how articles are formatted. As long as you don’t click on the “submit” button, you need not worry about inadvertently changing something.
Before creating a new article, it is a good practice to make sure that one doesn’t already exist by using the search function (left-hand side column). If you know that the article doesn’t exist, you may use the “create or find page” field (also on the left). If the page exists, you will be taken to it; otherwise, you will be taken to a blank page where you can then click on the “edit” tab to compose the article. Please note, though, that entering a page name will turn up an article only if the article’s page name is an exact match. The search, on the other hand, will do a general search.
Please note that the “Image Upload” function is for making images or graphics available to wiki articles and is not a method of creating a wiki article.
This wiki uses special syntax in order to handle various types of content. There is syntax associated with instructing the wiki about what category the article belongs to, and syntax for text formatting. The so-called markup language used for text formating is called “textile”1.
You can learn more about Textile at the author’s web site, which allows you to experiment with the various text formatting strings, or you could also look at or play with the try me page.
Before creating your first article, please read the information on this wiki’s article style conventions. Also, article editing contains some information on special fields used when revising an existing piece.
1 http://en.wikipedia.org/wiki/Textile_(markup_language)
]]>The triangle test is a discriminative method with many uses in sensory science including:
During a triangle test, a panelist is presented with one different and two alike samples. If possible, all three samples should be presented to the panelist at once, and the panelist should be instructed to taste the samples from left to right. The six possible order combinations should be randomized across panelists. For samples A and B, the six possible order combinations are: AAB, ABA, BAA, BBA, BAB, and ABB. The panelist is instructed to identify the odd sample and record his answer.
For evaluation with the chi-square distribution, use: X2=Σ (|O-E|)2/E, where O=observed and E=expected. To determine the number of expected correct answers, multiply the chance of choosing a correct answer by chance by the total number of panelists. In a triangle test, the probability of a correct answer by chance is 1/3. The probability of choosing an incorrect answer by chance is 2/3.
For example, a baking company recently reformulated their famous peanut butter cookie in order to reduce costs. The company wished to know if the reformulation was identical to the original. The researchers administered a triangle test to a panel of 60 tasters. The panel obtained 24 correct answers. For this problem:
H0: A=B Ha: A and B are not the same
A=reformulated product
B=original product
Oc=observed number of correct responses=24
Ec=n(1/3)=(60)(1/3)=20
OI=observed number of incorrect responses=60-24=36
EI=n(2/3)=(60)(2/3)=40
α=risk of a Type I error=0.05
From a chi-square distribution chart, X21, 0.05=3.84
Since X2=0.48<3.84, we fail to reject the null hypothesis and conclude that there is no significant difference between samples A and B. In other words, the reformulated cookie is not significantly different from the original cookie.
A Duo-Trio Test is an overall difference test which will determine whether or not a sensory difference exists between two samples. This method is particularly useful:
The Duo-trio test is statistically inefficient compared with the Triangle test because the chance of obtaining a correct result by guessing is 1 in 2. On the other hand, the test is simple and easily understood. Compared with the Paired Comparison test, it has the advantage that a reference sample is presented which avoids confusion with respect to what constitutes a difference, but a disadvantage is that three samples, rather than two, must be tasted.
Present to each subject an identified reference sample, followed by two coded samples, one of which matches the reference sample. Ask subjects to indicate which coded sample matches the reference. Count the number of correct replies and refer to a table for interpretation.
All should be familiar with the format, task, and evaluation procedure for the Duo-Trio Test. An orientation session is recommended prior to the actual test to familiarize subjects with the test procedures and product characteristics.
As a general rule, the minimum is 16 subjects, but for less than 28, the beta-error is high. Discrimination is much improved if 32, 40, or a larger number can be employed.
Offer samples simultaneously, if possible, or else present samples sequentially. Prepare equal numbers of the possible combinations (control and tests) and allocate the sets at random among the subjects. Space for several Duo-trio tests may be provided on the scoresheet, but do not ask supplementary questions (e.g., the degree or type of difference or the subject’s preference) as the subject’s choice of matching sample may bias his or her response to these additional questions. Count the number of correct responses and the total number of responses. Do not count “no difference” responses; subjects must guess if in doubt.
1 Meilgaard, M., G. V. Civille and B. T. Carr (2007). Sensory Evaluation Techniques, 4th Ed. New Boca Raton, FL: CRC Press, 6: 72 – 79.
]]>Is a broad category that encompasses all of the various methods used in the field of sensory evaluation and consumer research.
]]>Cluster analysis comprises a set of statistical techniques that aim to group “objects” into homogenous subsets. The objects can be people or products. For example, cluster analysis can be used to segment people (consumers) into subsets based on their liking ratings for a set of products. Such consumer segmentation is an essential step in preference mapping1, where the goal is to understand drivers of consumer liking, and cluster analysis is used to summarize differences among consumers in their likes and dislikes. Cluster analysis can also be used to cluster products instead of people, in an effort to identify groups of similar products, for example on the basis of trained panel sensory evaluations.
There are two common types of clustering methods: hierarchical methods and partitioning methods2,3. In the most common form of hierarchical clustering, the computation starts by assigning each object to its own cluster. Then, in a process of agglomeration, the most “similar” objects (in a statistical sense) are combined to form a new cluster, followed, in subsequent steps, by successively less similar objects. At each step in the hierarchy, the number of clusters decreases, until at the end all objects are combined into one large cluster. Hierarchical clustering programs typically produce a dendogram, a tree-like graphic representation of which objects are combined at each stage of agglomeration. The major limbs of the dendogram identify the major object groupings.
In the partitioning approach, the researcher must first specify the number of clusters that he/she is interested in. Objects are initially assigned to clusters on a random basis or on the basis of some prior knowledge or analysis. Using an iterative algorithm, the program reassigns each object to clusters until no further improvement in within-cluster homogeneity is achieved. The analysis is repeated for different numbers of clusters of interest to the researcher.
Among the several variants of agglomerative clustering, Ward’s method is one of the most popular and can be found in many statistical software packages. It is frequently used for consumer segmentation in sensory research. K-means is a widely used partitioning algorithm and is also available in many statistical software packages.
Each of the two methods of clustering (and each variation of the two methods) uses a different rule for forming clusters and therefore can lead to different answers when applied to the same set of data. Since all rules are essentially ad hoc, it is not possible to say in principle whether one approach is more correct than another. MacFie (2007)1 discusses the strengths and weaknesses of different agglomerative hierarchical methods in the context of preference mapping. Wajrock et al. (2008)4 compare hierarchical to partitioning methods for clustering.
None of the clustering approaches described above provide a definitive answer to the question of how many clusters exist. In practice, it is up to the researcher using these approaches to decide how many clusters make sense, considering the interpretability of the results and the number of objects per cluster that result from the analysis. For example, in segmenting a sample of 200 consumers based on their overall liking ratings for a set of products, a researcher might examine the results of analyses based on two to five clusters, and select the solution that is the most meaningful in light of the research objectives and that has sufficient objects (people) per cluster to be actionable.
Recently, latent class models5,6 have been proposed as alternatives to traditional hierarchical and partitioning methods. Latent class models can be viewed as probabilistic extensions of k-means7, where objects are assigned to a cluster based on the probability of belonging to that cluster given the observed data. Latent class models offer several potential advantages over traditional methods: they are based on a statistical model (the mixture of underlying probability distributions); they provide various diagnostics that are useful in determining the number of clusters; and they allow for “fuzzy” cluster assignment, where objects can belong to more than one cluster (with different probabilities). Applications of latent class cluster models in sensory consumer research are described by Popper, Kroll & Magidson (2004)8, Séménou et al. 20079, and Meullenet, Xiong & Findlay (2007)10.
Other approaches to cluster analysis include models that utilize external data in the process of clustering. For example, in a preference mapping application, such models can be used to cluster consumers’ liking ratings in a way that optimizes the predictability of liking ratings from trained panel evaluations (the external data). Cleaver & Wedel (2001)11 provide an example of a latent class regression model applied to sensory and consumer data. A different approach to utilizing external data was developed by Vigneau & Qannari (2002)12 and applied in a preference mapping context.
1 MacFie, H. (2007). Preference mapping and food product development. In H. MacFie (ed.), Consumer-Led Product Development. Cambridge, UK, Woodhead Publishing, pp. 551-592.
2 Jacobsen, T. & Gunderson, R.W. (1986). Applied cluster analysis. In J.R. Piggott (ed.), Statistical Procedures in Food Research. New York, NY, Elsevier Science Publishing, pp. 361-408.
3 Myers, J.H. & Mullet, G. M. (2003). Managerial Applications of Multivariate Analysis in Marketing. Chicago, IL, American Marketing Association, pp.238-304.
4 Wajrock, S., Antille, N., Rytz, A., Pineau, N. & Hager, C. (2008). Partitioning methods outperform hierarchical methods for clustering consumers in preference mapping. Food Quality and Preference, 19, 662-669.
5 Courcoux,, P., & Chavanne, P.C. (2001). Preference mapping using a latent class vector model. Food Quality and Preference, 12, 369-372.
6 Magidson, J. & Vermunt, J.K. (2004). Latent Class Models. In D. Kaplan (ed.), The Sage Handbook of Quantitative Methodology for the Social Sciences. Thousand Oaks, CA, Sage Publications, pp. 175-198.
7 Magidson, J. & Vermunt, J. (2002). Data use: latent class modeling as a probabilistic extension of k-means clustering. Quirk’s Marketing Research Review, 20 (March), 77-80.
8 Popper, R., Kroll, J., & Magidson, J. (2004). Application of latent class models to food product development: A case study. Sawtooth Conference Proceedings, 155-170.
9 Meullenet, J-F., Xiong, R., & Findlay, C.J. (2007). Multivariate and Probabilistic Analyses of Sensory Science Problems. Oxford, UK, Blackwell Publishing, pp. 111-127.
10 Séménou, M., Courcoux, P., Cardinal, M., Nicod, H., & Ouisse, A. (2007). Preference study using a latent class approach. Analysis of European preferences for smoked salmon. Food Quality and Preference, 18, 720-728.
11 Cleaver, G. & Wedel, M. (2001). Identifying random-scoring respondents in sensory research using finite mixture regression models. Food Quality and Preference, 12, 373-384.
12 Vigneau, E. & Qannari, E.M. (2002). Segmentation of consumers taking account of external data. A clustering of variables approach. Food Quality and Preference, 13, 515-521.
]]>When creating a new wiki article, please following the style guidelines described below. Having a consistent style is very important for readability.
First and foremost, please choose an article page name fairly carefully, as this will be the system identifier of the article. The page name should be as short as possible without unduly sacrificing specificity. For example, an article on palate cleansing could be named “palate cleansing” if it was a general treatment of the topic, but if it was specific to spicy foods and there was already an article that dealt with, say, lingering sweeteners, you may then want to include “spicy” in the page name.
Also note that when a wiki article refers to another wiki article, it will do so using the page name as the text of the (internal) hyperlink. Hence, an article page name should be something that can be logically and readily used in a sentence.
To help organize the wiki, articles are assigned to categories. For example, this article was put under the “help” category. There are also several articles that are under “basic taste.” If the article can be placed within an existing category, it should be. But it may certainly happen that a new category is the first of its kind.
There is a special syntax used to assign an article to a category. The word “Category” and the name of the category must be separated by a colon and enclosed with a set of matching double brackets. Note that there are no spaces either before or after the colon character. For example, this article was categorized by using the following as the first line:
[[Category:Help] ]Please set the name of the article, which can be longer than the article page name, in larger text using a heading style of “h1”. To do this, simply type “h1” followed by a period, and then a single space before the title text.
Material already in the wiki should never be copied and pasted into another article, but, instead, it should simply be referenced. To reference another page within the wiki simply enclose the article name within matching double brackets. For example, suppose we wanted to refer back the general, main, help page. In the article, such a reference would look like this: help. When editing, this reference would be entered as follows:
[[help]]When an article needs to reference a web page external to the wiki — that is, to provide a standard hyperlink — the syntax for that is somewhat more involved. Suppose that you wanted to reference wikipedia’s article on wiki’s and have the hyperlink formatted as: what is a wiki?. This would be accomplished as follows:
"what is a wiki?":http://en.wikipedia.org/wiki/Wiki
Notice that the hyperlink text is first and enclosed in double quotes, followed by a colon, and then finally the full URL.
Whenever possible, if statements are made that are not common knowledge, or a quote is used, a reference needs to be given. In order to avoid hard-to-read, cluttered text, references can be included through the use of footnotes. A footnote can be included by just enclosing its sequential number in matching single brackets. Later, at the end of the article, the footnote text is specified with the “fn” command followed by a number, immediately next to the text.
For example, suppose one made the following quote:
You can, for example, never foretell what any one man will do, but you can say with precision what an average number will be up to1.
The footnote reference was entered as follows:
... average number will be up to[1].
Later, the actually footnote might look like:
1 Sherlock Holmes in “The Sign of Four,” 1890.
And was entered as follows:
fn1. Sherlock Holmes in “The Sign of Four,” 1890.
If a sentence has multiple footnotes, the footnotes should be separated with super-scripted commas as follows:
...end of sentence.[5][^,^][6]
and would be formatted as follows:
And the end of the article, in the footnote section, there needs to be a blank line between each footnote (the “fn” piece).
If needed by the article, a graphics image (whether a chart, equation, etc.) can be used by first uploading it to the wiki images area using the “File Upload” menu option. The system guides you through the process, and ends by displaying the appropriate way to reference the image in the article. When using images, please be sure to save the file using the smallest size reasonable.
Because the society site has forums built-in, any discussions of article content should be done using the article forum. If an article does not have a discussion ongoing, there will be a menu tab that says “start a discussion”, and if there is a discussion already started, it will say “discuss article”.
]]>