Labeled scales (e.g., category, Likert, VAS) and invalid across-group comparisons: what we have learned from genetic variation in taste

It is argued that labels used as standards often refer to experiences of different absolute intensity to the groups of interest making the comparisons invalid, and data is presented showing this assumption is not true.

Abstract

Direct comparisons of sensory or hedonic perceived intensities across individuals are impossible since we cannot share experiences. However, experiences can be compared indirectly if we can identify a standard assumed to be equal, on average, to groups compared. Sensory standards (i.e., magnitude matching) have proved very useful. Intensity adjectives/adverbs have also been used as standards (e.g., This tastes very strong to me; is it very strong to you?). We argue these labels often refer to experiences of different absolute intensity to the groups of interest making the comparisons invalid. An early solution rested on the assumption that the maximum intensity perceivable was equal for all sensory domains and individuals. Comparisons were then possible because all judgments could be made on a zero to maximum scale. We present data showing this assumption is not true.