RESUMEN
Check-All-That-Apply (CATA) is a very popular tool for rapid sensory profiling in consumer research. Yet, consumers' dietary habits, such as consumption frequency, has been neglected in regard to the impact, if any, on CATA question responses. The present work aimed to fill this gap, by focusing on the effect of different consumption frequencies with four studies (N = 686). Two categories of drink products were involved, chrysanthemum tea and instant coffee. For each category, two sets of products were prepared to constitute two levels of difference between samples (larger vs. smaller). Consumers were classified into high- and low-consumption groups according to their consumption frequencies of each focal product; the size of these subgroups ranged from 54 to 130. Overall, the two groups did not produce large discrepancies from each other when constructing sensory profiling of the tested samples with CATA questions as well as the stability of sample configurations. However, there were some nuances between them. In the evaluation of chrysanthemum tea samples, the higher-consumption group of consumers presented better discrimination than the lower-consumption group while this was reversed for coffee samples. This might be mainly attributed to the fact infrequent consumers were more sensitive to such negative attributes as "strong chrysanthemum smell for chrysanthemum samples, earthy for coffee samples". Further, despite the higher stability obtained with larger sizes of difference between samples in Studies 3 and 4, in general, the effect of size of differences within sample sets was not significant between different consumption groups. To more effectively contribute to refinement of methodological guidelines for CATA questions, more validation work is needed.