Comparison of methods for generating sensory vocabulary with consumers: A case study with two types of satiating foods

- Three methods for generating the consumers' vocabulary were compared.
- Convergent validity was observed.
- Repertory grid and individual sample description provided the richest vocabulary.

Understanding consumers' sensory vocabulary can help to identify the most relevant characteristics for liking and reduce the differences with descriptions provided by trained panels. The present study compared three methods for generating sensory vocabulary with consumers: repertory grid (RG), comparison of the sample set (CW), and individual sample description (ID). A consumer study was carried out with two groups of 90 participants, randomly divided into 3 subgroups (n = 30) which evaluated one of the two types of satiating product (chocolate-flavoured shakes and biscuits) using one of the three methods to generate sensory vocabulary. The methods differed in their ability to encourage consumers to verbalize their sensory perception in detail, and also in the total number of terms generated and the frequency with which consumers generated some terms. The lowest number of terms, at both individual and aggregate level, was elicited when consumers described similarities and differences in the whole sample set. However, the most frequent terms were similar in the three methods, indicating their convergent validity. Results from the present work suggest that ID and RG are preferable to CW. Considering practical aspects related to their implementation, ID shows advantages over RG for identifying the consumers' vocabulary without seriously compromising data quality. Texture attributes, which have been related to satiating capacity, were frequently mentioned to describe the evaluated products. Terms related to viscosity were the most frequently elicited, whereas varied textural features were mentioned for biscuits.