Assessment of Global and Individual Reproducibility of Projective Mapping with Consumers

Abstract

Abstract The popularity of projective mapping with consumers for sensory characterization has markedly increased in the last 5 years. To have confidence in this methodology, it is necessary to ensure that a similar product profile would emerge if the test was repeated. Also, deciding whether the study should be replicated or not is a key issue in test implementation. In this context, the aim of the present work was to evaluate global and individual reproducibility of projective mapping for sensory characterization with consumers and to evaluate the influence of the size of difference among samples. Six consumer studies were conducted using a test–retest paradigm. In each study, responses from the same group of consumers to the same sample set in two different sessions were compared. Across the six studies, individual reproducibility tended to be low. However, the RV coefficients of consensus sample configurations between sessions were higher than 0.75, suggesting that test–retest reproducibility of projective mapping with consumers proved to be relatively high. Practical Applications The present work provides evidence of the reproducibility of projective mapping for sensory characterization with consumers. Although sample configurations were stable, some differences in conclusions regarding similarities and differences among samples were identified between sessions. This indicates that care must be taken when relying on results of projective mapping with consumers obtained without the use of replicates, particularly when working with sample sets with small differences. Results from the present work showed that stability indices of sample configurations based on bootstrapping resampling approaches were related to global reproducibility. These indices could be useful to decide whether or not it is necessary to replicate projective mapping in order to ensure that conclusions regarding similarities and differences among samples would be repeatedly identified. This is of particular interest, considering the difficulty of asking consumers to attend separate sessions.