Anxiety in verbal behavior: An intercorrelational study.

Publisher Website
Google Scholar
Abstract
Several objective measures of anxiety in verbal behavior have been proposed in the last 3 decades. Their claims to validity are relatively weak, generally resting upon their apparent reasonableness. If they appear to reflect the same "inner state" and so yield highly correlated measurements on the same behavior samples, their claims would be stronger. They might be said to show concurrent validity. Ten-minute recorded samples from the therapy sessions of 15 hospitalized male mental patients were studied. Eight purported measures of anxiety were applied to each patient's verbal response in the 15 protocols. An intercorrelation matrix was computed for each of the 15 protocols to study the amount of individual differences, and then the matrix of median intercorrelations was derived. The average results have a very large sampling error over persons. The average interquartile range for the median values was .45. Thus, there are marked individual differences in the level of intercorrelation among the various verbal anxiety measures. This implies that different measures may be valid for different persons and that what measurement values indicate anxiety or nonanxiety may also be idiosyncratic. (PsycINFO Database Record (c) 2006 APA, all rights reserved)

This publication has 0 references indexed in Scilit: