Search

The R‐index is a sound methodology, which has been introduced, developed, and used in the sensory and consumer fields for several decades. The R‐index methodology can be used both to measure and to test sensory difference and consumer preference. This paper provides an R code for calculation of the R‐index critical value for difference testing, which is a supplement to or an alternative for the published table in Bi & O'Mahony (2007) (Journal of Sensory Studies, 22, 713–720). Practical applications: The R code provided in the short contributions overcomes the limitation of the previously published table of R‐index critical values in Bi & O'Mahony (2007) (Journal of Sensory Studies, 22, 713–720) with its limited number of sample sizes and required equal size for the control sample and test sample. [ABSTRACT FROM AUTHOR]

In recent sensory literature, Bi and Kuesten (2015a. Food Quality and Preference, 43, 47-52; 2015b. Journal of Sensory Studies, 30, 461-471) have proposed the 'M + N' classification method with larger M and N (>3) for discrimination testing based on Fisher's exact test. Industry studies using the method are shared in this article to illustrate application of the method. The tests and measures are taken for non-fatiguing visual and manual inspections conducted across a variety of product categories. Further, within the examples provided, a validation between labs is documented and results of the method are compared to results from other methods (2-Alternative Choice and 2-Alternative Forced Choice) based on common conventional sensory measurement Thurstonian d′. Estimations are given of Intraclass Correlation Coefficient and Cronbach's coefficient alpha for evaluation of panel performance in the test using the 'M + N' classification method with larger M and N (only for M = N > 3). Practical applications Sorting methods are popular and often used for rapid investigation of product quality, product differences, or consumer perception and are used to help guide product maintenance and/or development efforts. The 'M + N' classification method with larger M and N, a type of sorting technique, can be applied for sensory discrimination and perceptual tests with generation of various indices, including Thurstonian discriminal distance delta (or d′) thereby offering the ability to compare results with other sensory discrimination methods. [ABSTRACT FROM AUTHOR]

The A- Not A with reminder ( A- Not AR) is a relatively new method in the sensory literature. This paper derives and demonstrates the function illustrating the relationship between the area under the receiver operating characteristic ( ROC) curve and d′ for the A- Not AR with the differencing strategy. This function shows that the area under an ROC curve for the A- Not AR is equal to the maximum proportion of correct responses in an A- Not A. This theoretic function can be used to estimate d′ and its variance from the ratings of the A- Not AR. A simulation study shows that the nonparametric estimation based on this function is close to those obtained by using the maximum likelihood estimation. The estimation of the variance of d′ based on the delta method is close to that obtained by using the bootstrap method. R codes are provided for the estimations of d′ and its variance as well as simulations. Practical Applications The A- Not AR is a variation of the conventional A- Not A method, which is used in the sensory field for discrimination testing. This method is particularly useful when it is difficult to have an adequate familiarization procedure for the test samples for panelists before a test. This approach also offers the potential for applying such a method to problems like hedonic, purchase intent, and consumer concept measures in addition to other measures in the sensory and consumer science field. This paper provides a new nonparametric method and R codes for estimating d′ and its variance from the ratings of the A- Not AR. [ABSTRACT FROM AUTHOR]

ABSTRACT This article attempts to deliver the following message to the researchers and practitioners in the sensory field. (1) Theoretically, drivers of consumer liking is based on relative importance of explanatory variables in a linear model. The problem is complicated when the variables involve linear dependence, which is the common situation in sensory and consumer data. (2) The commonly used methodologies, e.g., conjoint analysis, preference mapping and Kano's model, have serious limitations for determination of relative importance of correlated attributes and identification of drivers of consumer liking. (3) The conventional statistics, e.g., correlation coefficient, standard regression coefficient and P values of tests for regression parameters, etc., are inadequate and invalid measures of relative importance of correlated attributes. (4) There are three state-of-the-art methods for determination of relative importance of correlated attributes. They are the Lindeman, Merenda and Gold's method, Breiman's Random Forest and Johnson's relative weight. This article also provides statistical background and almost exhaustive main references on the topic of relative importance of variables scattered in various academic journals in different fields. The information will help the sensometricians and researchers with more statistical knowledge to embrace the mainstream of the research on the topic and to pursue advanced methods for drivers of consumer liking. PRACTICAL APPLICATIONS This article reviews some new methods for determination of relative importance of correlated explanatory variables to response variable in a regression model. The methods can be used for identification of drivers of consumer liking. The article also provides the sources of the corresponding computer packages and codes implementing the new methods. The packages and codes are freely available and easy to use. The R packages 'relaimpo' for the LMG method, 'randomForest' and 'party' for the original and modified Breiman's Random Forest method are available at . The R or S-Plus code 'johnson' for Johnson's relative weight is available from the online supplementary Appendix S1 of this article. [ABSTRACT FROM AUTHOR]

ABSTRACT This paper introduces and utilizes three new methods, which appear in statistical and machine learning literature, to identify drivers of consumer overall liking. These methods include the Lindeman, Merenda and Gold (LMG) method, Breiman's Random Forests (RF) and Johnson's relative weight. The methods provide valid measures of relative importances of attributes to consumer overall liking in the situation of multicollinearity. The methods are flexible and can be used for both aggregated data (averages) of multiple products and raw data of one or more products. Breiman's RF and Johnson's methods can be used for both the data sets with more observations than variables and the data set with more variables than observations, which is the typical situation in sensory panel descriptive analysis data. The LMG and Breiman's RF methods are applicable to both continuous attributes and categorical (including binary) attributes. Numerical examples are provided to illustrate the applications of the methods using freely available R programs. PRACTICAL APPLICATIONS Identification of drivers of consumer overall liking is a main objective of sensory and consumer study. The paper provides three new approaches to the aim. With the freely available R programs, practitioners can easily and quickly conduct the analysis. [ABSTRACT FROM AUTHOR]

This paper discusses similarity tests using forced‐choice methods in terms of Thurstonian discriminal distance, d′, a measure of pure sensory difference or similarity. Critical values are derived for the similarity tests using the 2‐alternative forced choice (AFC), the duo‐trio, the 3‐AFC and the triangle methods in terms of d′. Power analysis is conducted for the similarity tests using the four forced‐choice methods. Three approaches are used to estimate sample sizes for the tests. The approaches are based, respectively, on exact binomial distribution, normal approximation with a continuity correction and Monte Carlo simulation. S‐Plus and R codes were developed and provided for calculation and estimation of the critical values, testing powers and sample sizes for the tests. PRACTICAL APPLICATIONS: Similarity testing has wide application perspective in sensory and consumer research. It should be used when similarity rather than difference is of interest. The similarity tests discussed in the paper can be used to infer if a difference, in terms of d′, between, e.g., a prototype and a control, is smaller than a similarity limit, i.e., a specified allowed difference. [ABSTRACT FROM AUTHOR]

The four-alternative forced choice (4-AFC) is one of a family of m-AFC tests used in Psychology. Based on a detection-theoretic model, for a given d′ it is slightly more powerful than the 3-AFC and 2-AFC. Although its potential use in sensory food science is limited, there are occasions when it will be used. Accordingly, tables of d′ values for given proportions of correct responses (pc) along with corresponding values of B, for computing variances, are given in this paper. Power comparisons are also made between 2-AFC, 3-AFC and 4-AFC tests. PRACTICAL APPLICATIONS For sensory difference testing, the choice of test will depend on several factors such as statistical power as well as various physiological and cognitive effects. Because of the wide variation in the sensory effects that food and personal product stimuli can have on judges, it is important to have a wide variety of test methods available. Therefore, even less common tests need to be considered. The 4-AFC can be used on occasions and it is certainly suitable for visual judgments. Accordingly, for this method, it is appropriate to have tables of d′ and B for computing variances. [ABSTRACT FROM AUTHOR]

Conventional drivers of liking analysis was extended with a time dimension into temporal drivers of liking (TDOL) based on functional data analysis methodology and non-additive models for multiple-attribute time-intensity (MATI) data. The non-additive models, which consider both direct effects and interaction effects of attributes to consumer overall liking, include Choquet integral and fuzzy measure in the multi-criteria decision-making, and linear regression based on variance decomposition. Dynamics of TDOL, i.e., the derivatives of the relative importance functional curves were also explored. Well-established R packages ‘fda’, ‘kappalab’ and ‘relaimpo’ were used in the paper for developing TDOL. Applied use of these methods shows that the relative importance of MATI curves offers insights for understanding the temporal aspects of consumer liking for fruit chews. [ABSTRACT FROM AUTHOR]