Measuring tolerance of ambiguity: Item polarity, dimensionality, and criterion validity

Research involving tolerance of ambiguity (TA) has produced inconsistent findings for half a century. Some of this inconsistency may be due to the psychometric shortcomings of the TA measures. In this investigation, the Budner (1962) and McLain (1993) TA scales were examined using confirmatory factor analyses. Both scales were tested against a single-factor model, a two-factor model, a two-factor correlated model, and a two-method/single-trait model with method defined by item polarity (positive and negative). Results demonstrated that the two-method/single trait model was the best fit for both scales. However, whereas the McLain scale demonstrated adequate psychometric integrity, the Budner items failed to load sufficiently on the trait factor suggesting a weakness in validity. Correlations with verbal ability and inconsistent regression results with respect to the Budner scale support the choice of the McLain scale in future work. (C) 2010 Elsevier Masson SAS. All rights reserved.