Context Effects and the Communicative Functions of Quantifiers: Implications for Their Use in Attitude Research
Many descriptions of states of affairs or beliefs are quantified. Quantification in its most general sense involves the use of numbers and proportions (e.g., 3,2%) and more typically a whole host of natural language expressions, including quantifiers (a few, some, most), quantifying frequency adverbs (sometimes, often, usually), and probability expressions (possibly, likely), among others. It may be the case that when they are used, such expressions play the role of numerical descriptions without requiring the exactness of numbers. Sometimes this is true, and such an approach has led to research on how the mapping between verbal descriptions and decisions under risk, particularly using probability or likelihood expressions (e.g., Budescu, Weinberg, & Wallsten, 1988; Rapoport, Wallsten, Erev, & Cohen, in press). The results of these studies show that numerical information is more accurate than that information conveyed by a verbal expression and is consistent with the view that quantifiers do not denote single values but instead specify ranges, which are describable as membership functions on probability or frequency continua. Obviously, the same argument can be made for the importance of ranges with quantifying determiners (see Moxey, 1986): A word such as “many” can be thought of as denoting a range that the user has in mind, for instance.
Unable to display preview. Download preview PDF.