Abstract
A previous study reported that the correlation between attitude scores on Thurstone and semantic differential scales was greater when issue saliency was low than when issue saliency was high. The present study yielded the opposite results on two different issues using two different sets of attitude scales. A more detailed analysis, however, indicated that the low correlations in both studies were probably produced by a restricted range of scores on one or both scales. Contrary to the previous study's warning, the present results indicate that the semantic differential scale is no more vulnerable to changes in issue saliency than are other widely used measures of attitude.
Get full access to this article
View all access options for this article.
