![xlstat jar xlstat jar](https://cdn.xlstat.com/media/feature/0001/01/thumb_197_feature_large.png)
You need to create one mean score for each of your groups. This can be done in a raw Excel data file or another data package. Now you will need to look at how each of these groups answered the overall appeal question. Next you will need to calculate the size of each of these two groups, by dividing the number of respondents in each group (A,B,C) by the total number of respondents that answered your study. Next those who code ‘just about right’ (3) will be group B, and those who selected either ‘not at all sweet enough’ (1) and ‘not quite sweet enough’ (2) will be group C. So those who selected ‘a little too sweet’ (4) and ‘much too sweet’ (5) we will call group ‘A’. Looking at each of your JAR scales, you need to group consumers who rated the product too sweet, or not sweet enough. For example: sweetness, thickness, colour and strength of smell. You must also have a series of JAR scales against which to measure the influence. How much do you like the product overall, where 5 is like a lot and 1 is dislike a lot? This could also be asked on a 7 or 10 point scale if preferred. For this we use Penalty Analysis.įirstly, you need to have a question that measures the consumer’s overall appeal of the product e.g. This will tell you whether this attribute is critical to optimise, or of secondary importance. The next stage is to assess the influence that this attribute is having on your overall appeal, and the penalty encountered when you under-perform. A mean close to zero will illustrate whether you are in line with consumer expectations for this attribute, but it is important to look whether this mean is derived from polarised or unanimous scoring. When answering the questions, the respondent can only select one of the five answers, making this a single code question.Īt the end of the study, you will be able to create a mean score on this data, finding your average position in relation to being just about right.
![xlstat jar xlstat jar](https://image.slidesharecdn.com/finallecture2withoutvoiceppt-170805060522/95/lecture-2-latentmanifestobserved-variables-using-in-sem-analysis-wwwstatsworkcom-8-638.jpg)
To quantify this, we create five unique positions on a 5-point scale ranging from much too sweet, to not at all sweet enough for example (Figure 1). For example, the sweetness of a chocolate bar could be Just About Right, or it could be either too sweet, or not sweet enough. The scale assumes that there is an ideal position for an attribute, and the possibility of being over or under the ideal.
![xlstat jar xlstat jar](https://www.statcon.de/produkte/produktbeschreibungen/xlstat_sensory_general_dialog_box_prefmap.png)
Just About Right (JAR) scales are a commonly used question format when trying to identify the performance of a product or experience against a certain attribute.