XLSTAT: Multivariate analysis and clustering on JAR (Just-About-Right) data in Excel

 

Background

Clustering in Just About Right (JAR) analysis is used to group consumers with similar perceptions of a product's attributes. JAR scaling measures whether a specific attribute is at an ideal level according to consumers.

 

Dataset for Example

The dataset provides valuable information on 72 assessors evaluating 8 different cheeses based on 9 JAR attributes.

 

Setting Up

• Open XLSTAT.

• Click on the Sensory/ Rapid tasks menu and select the JAR Multivariate analysis and clustering feature.

• The dialog box pops up.

In the General tab:

• Select the just-about-right data corresponding to the descriptors.

• Identify the  scale corresponding to the JAR data.

• Select the data corresponding to the product identifiers.

• Select the data corresponding to the assessor identifiers.

• Choose between the Explanatory analysis (CATATIS) and Clustering analysis (CLUSCATA). 

In the Options tab:

• Enter beta, the parameter for agreement between JAR and other parameter as follows:

• Click on the OK button to launch the computations.

• The results are displayed on a new Excel sheet.

 

Interpreting the Results

Multivariate analysis of JAR data

The products/assessors table is displayed first, then all the results for the CATATIS method adapted to JAR data. For example, we can see that product R is characterized by too little texture and consistency and is too creamy.

For a better understanding of the results, please see Analysis of CATA data with CATATIS in Excel.

 

JAR data clustering

The products/assessors table is displayed first, then all the results for the CLUSCATA method adapted to JAR data.
Three clusters of assessors could be considered here.

For a better understanding of the results, please see Clustering subjects in a CATA task by means of CLUSCATA.