Product spaces derived from projective mapping and CATA questions: Influence of replicated assessments and increased number of study participants

Abstract Methodological research is needed to support the continued popularity of sensory product characterization by consumers. The present research considers stability and reliability of such product spaces. Using data from 13 studies ( n = 1332) involving check‐all‐that‐apply questions (CATA) and projective mapping tasks replicated assessments and increasing the sample size are evaluated for their impact on the product spaces derived from a base group of consumers. In each study, a group of consumers (41–95 participants) completed a sensory characterization task twice under identical conditions (“replication” condition), and additional consumers (49–93 participants) completed the task (“increased sample size” condition). The stability of product spaces was evaluated using a bootstrapping resampling approach. Both experimental conditions had a small but positive impact as evidenced by small increases in average RV coefficients across simulations. Additionally, in studies using projective mapping, the confidence ellipses around sample positions were smaller than those established in product spaces with the base group of consumers. This pointed to increased sample discrimination. Practical applications This research reveals that replication and increased sample size by inclusion of additional consumers can be used to increase stability of product spaces from sensory characterization based on projective mapping and CATA questions. Stability in product spaces underpins reliability of the research, and by extension internal validity.