RANGE SENSITIVITY OF ATTRIBUTE WEIGHT IN MULTIATTRIBUTE VALUE MODELS

In decision analysis, multiattribute value functions are normalized re lative to the best and worst outcomes in the local decision context. W ith this normalization, attribute weights (scaling constants) should v ary as a function of the range of outcomes on each attribute in the lo cal context. Other things being equal, the greater the range of outcom es for attribute X, the greater the weight for attribute X should be. Two experiments tested this normative range-sensitivity principle for three weight assessment methods: direct importance weights, trade-off weights, and swing weights. Weights elicited using the direct importan ce weight method were range-insensitive, contrary to the standard norm ative model. Weights elicited using the swing weight and trade-off met hods both displayed statistically significant range sensitivity, altho ugh both methods were less range-sensitive than predicted by the norma tive model. The studies revealed another bias as well. Trade-off judgm ents gave greater weight to the most important attribute than did dire ct importance ratings or swing weight assessments. (C) 1995 Academic P ress, Inc.