| Three studies focused on development and validation of a reliable and valid pencil-and-paper measure of ambiguity intolerance. Study 1 revealed psychometric problems in the Budner (1962) and Norton (1975) scales of ambiguity intolerance, indicating that these scales may not be reliable or valid measures of the construct. A new 54-item measure of ambiguity intolerance was developed in Study 2. This measure was proved to be both internally consistent (alpha =.94) and stable (test-retest reliability, r =.84), and convergent and discriminant analyses suggested that the measure of ambiguity intolerance related in meaningful ways to measures of other constructs which share features with it. Construct validity experiments (Study 3) provided mixed support for the prediction that high ambiguity intolerant subjects would have greater difficulty in situations that were highly ambiguous than in less ambiguous situations when compared to low ambiguity intolerant subjects. On measures of visual perception, high ambiguity intolerant subjects had longer response latencies to high ambiguity stimuli than to low ambiguity stimuli, and as a group, high ambiguity intolerant subjects had longer response latencies than low ambiguity intolerant subjects. While the predicted interaction of ambiguity intolerance and stimulus ambiguity was not confirmed on response accuracy, high ambiguity intolerant subjects were more accurate on all stimuli than low ambiguity intolerant subjects. Considering accuracy and latency together, high ambiguity intolerant subjects had comparable rates of solutions-per-unit-time to low ambiguity intolerant subjects. The predicted interaction of ambiguity intolerance and scenario ambiguity in person perception failed to occur, possibly because scenario ambiguity was not successfully manipulated. However, the effects of ambiguity intolerance on latency and accuracy replicated some of those on the visual perception measures, where high ambiguity intolerant subjects were slower but more accurate than low ambiguity intolerant subjects. Again, the rate of solution-per-unit-time for high intolerant subjects was comparable to low ambiguity intolerant subjects. Finally, ambiguity intolerance was not associated with poorer performance on a test of abstract thinking. However, this appeared to be attributable to poor performance on this task by all subjects, over 50% of whom scored at levels indicative of organic impairment. Implications for ambiguity intolerance theory are discussed. |
