Many decisions require the application of subjective judgment. Two methods of modelling such decision behaviour by using fuzzy set theory are considered here. Yager (1 978) uses a method based on a maximin criterion, while Tsukamoto and Terano (1975) base their method on multivalued logic operations. In this paper sets of decision maker's evaluation of decision problems are compared with each of the two fuzzy set evaluations. The ability of the fuzzy set methods to replicate decision behaviour is measured by the rank correlation between orderings of alternatives obtained from the decision makers and those obtained from the. methods. Analysis of the experiment indicated that for the sample used, the methods generated very few correlations that were significantly greater than zero. These results appear to contradict claims for the methods made in the literature.