A comparison of RATA questions with descriptive analysis: Insights from three studies with complex/similar products

Consumer-based methods for sensory product characterization continue to gain popularity. Considering that past research is mostly limited to cases where samples are sensorially simple or markedly different, the aim of this study was to compare descriptive analysis (DA) by trained assessors with product characterizations from consumers using rate-all-that-apply (RATA) questions using samples that can be regarded as quite similar or quite complex. The results from three studies showed that sample differentiation was higher in DA than RATA and that information about sample similarities and differences tended to be dissimilar. The exception was Study 1 where samples were least complex. The discrepancy between the two types of panels was least evident for appearance and color attributes and basic tastes (sweet, sour, and salty), and in these instances, the correlation between attribute intensities was clearly linear (R-2 = 0.46). The discrepancy was highest for complex sensory terms and terms relating to specific odors/flavors. While past research suggests that it may be possible to improve the agreement between the two methods by providing consumers with short training (samples, terms, and rating scale), practitioners should consider that training of consumers may reduce the ecological validity of their responses. Practical applicationsCharacterization of the properties of products is a cornerstone activity for sensory professionals, and to achieve this task, several methods are available. This study contributes insight regarding the choice of descriptive analysis (DA) with trained assessors or rate-all-that-apply (RATA) questions with consumers. If the products are not sensorially complex, then product characterization by consumers provide the same insights as DA. However, this study shows that when samples are sensorially complex and/or similar, RATA is less discriminative than DA and provides information about sample similarities and differences that is not the same as that generated by DA. Practitioners are advised to select sensory description methodology based on the objectives of their research, and when an understanding of consumers' perceptions of products is desired, A consumer-based methodology should be used.