Analysis of data from CATA questions

Data from CATA questions consist of binary data which indicate if each consumer has selected each of the terms for describing each of the samples included in the study. An example of a typical data matrix, for a study with five samples and 60 consumers, is shown in Table 11.1. In the matrix, each column corresponds to a term of the CATA question. Using this data matrix, different approaches can be used for data analysis, which will be discussed in the following sections. 

The relevance of each term included in the CATA question for describing each sample is determined by calculating its frequency of use. Data are usually summarized using contingency tables which contain the number of consumers who selected each term for describing each sample (Meyners et ah, 2013). Data can be displayed using counts or percentages, but the latter is more common. 

Cochran s Q test is used to evaluate if consumers detected significant differences among samples for each of the terms from the CATA question. Cochran s Q test is a 

Consumer Sample Sweet Sour. .. Bitter Thick 

1 indicates that the consumer selected the term for describing the sample, while 0 indicates that the term was not selected. Data shown without randomization of terms and samples to more clearly visualize the structure. 

---