Two Step Double-Faced Applicability Test
Introduction
In the 'double-faced applicability' test, each attribute is presented with a “double-faced” approach, featuring two descriptors (a pair of semantic-differential descriptors) separately presented in the questionnaires to represent both aspects of the attribute.
When performing the two-step rating for each attribute in the questionnaire, participants are directed to first indicate whether a specific aspect of one attribute applies by choosing either "Yes (does apply)" or "No (does not apply)."
Subsequently, they are required to provide a 3-point sureness rating, indicating their level of certainty regarding their initial Yes or No response. The same task is subsequently repeated for the attribute representing the opposite aspect.
Template Description
The template includes an initial screen, which can be used to provide instructions to the panellists.
The template includes two design sections. The first design section features three screens, which will be shown to all the products included in the design, focusing on the positive aspect of three attributes.
Within each screen, panellists are prompted to assess whether a specific attribute applies to the product. Subsequently, panellists are asked to indicate their sureness in the previous response.
Once the evaluation of positive attributes for all products is complete, panellists will be given a short break. This section can be utilized for additional instructions or to allow panellists to take a break before proceeding with the test.
After the break, panellists will evaluate the same products, but this time they will focus on the negative aspects of the attributes. For instance, if the initial evaluation included attributes like ''Shiny'', Not leaving residue, and Convenient to use, the second part will involve evaluating attributes such as Not shiny, Leaving residue, and Inconvenient to use.
Within each screen, a formula question type is integrated. The formula embedded in this template facilitates the computation of results, which will be presented in the raw data. The formula calculation is derived from the study by Kim, I. A., Hopkinson, A., van Hout, D., & Lee, H. S. (2017).
The gathered responses will be transformed into a 6-category single scale of values based on Signal Detection Theory (SDT). For instance, if a subject selects "Yes (does apply)" and "3 (very sure)" in a 3-point sureness rating, the responses will be stored as 6. Similarly, if subjects choose “No (does not apply)” and “3 (very sure)” the responses will be stored as 1 (Figure 1).
Figure 1: Data Matrix for Formula Calculation (Adapted from Kim et al., 2017).
The formula is derived from the question name currently added in the template. Should the question name for attributes be changed (e.g., from "Attribute 1" to "Shiny"), the formula must be adjusted accordingly. Replace every instance of "Attribute 1" in the formula with the word "Shiny," maintaining the exact spelling as it appears in the question name.
The positive and negative attributes in each block will be presented randomly within a subject to minimize primacy bias. To implement this randomization in EyeQuestion, a screen setting can be utilized. Within each screen, you can designate the screens you wish to randomize by assigning them to the same Random Group.
In this template, all screens featuring positive attributes belong to Random Group A, while screens with negative attributes belong to Random Group B.
The end screen of the template features a thank-you message, offering the opportunity to customize a final message for panellists.
Within the design settings, you can review how each panellist will receive the sample. Each set corresponds to a specific order of presentation, with panelists receiving the samples in a monadic sequential order, one product after another. For the following template, two products are included: however, if more products are wanted, you can add more by clicking the on Generate Design button and increasing the number of products.
Raw Data Representation
The raw data will be displayed, with each row corresponding to an individual judge and the assessed sample. The data will indicate the value selected by panellists for each attribute in response to the "Yes (does apply)" or "No (does not apply)" question, along with the associated sureness rating. Additionally, a third column will show the outcomes of the formula question type for each attribute.
References
- Kim, I. A., Hopkinson, A., van Hout, D., & Lee, H. S. (2017). A novel two-step rating-based ‘double-faced applicability test. Part 1: Its performance in sample discrimination in comparison to simple one-step applicability rating. Food Quality and Preference, 56, 189-200.
Related Articles
Two Out of Five
Introduction The Two Out of Five method is used to understand the presence or absence of a perceptible sensory difference or similarity between samples of two products. Panellists are presented with five samples and asked to identify the two samples ...Discrimination Analyses Overview
Overview Difference tests are the fundamental and most frequently used methods in sensory and consumer research. The most important questions difference tests must answer are: do various products differ from each other? And does the consumer detect ...Discrimination Test Settings - Pd and d' Analysis
Purpose Establish the power of a discrimination test given a set sample size or to calculate the sample size required to get a desired power. This can be done specifying the expected difference as a proportion of discriminators (Pd) or as a ...Two-Alternative Forced Choice (2-AFC)
Introduction The Two-Alternative Forced Choice (2-AFC) method, is a specific way of comparing two samples in sensory science. In this method, the only allowed difference between the two samples is a specified sensory characteristic. It works well ...Tetrad
Introduction The Tetrad method is a testing procedure designed to understand the presence of a perceptible sensory distinction between two products. Panelists are given four samples, with two being identical and two being different. The panellists ...