Doctor of Philosophy
James P. McCullough
Recent research has provided support for a multidimensional view of trait anxiety to supplant the former unidimensional approach. Unidimensional measures of general trait anxiety have been found to be inadequate as predictors of state anxiety reactions across a wide variety of situations. As such, they are poor measures of general trait anxiety. The present investigation was conducted to examine the possible utility of a single anxiety trait score, summed from the subscales of the Stimulus - Response Inventory of General Trait Anxiety (S-R GTA), a multidimensional measure of trait anxiety, in supplementing the ability of individual subscale scores to predict state anxiety responses to trait-congruent situations. It was hypothesized that subjects who scored at the same level on a particular subscale would differ in their responses to trait-congruent stress because of differences in general trait anxiety as measured by the total score on the S-R GTA. A second purpose of this study was to examine the possibility that different dimensions of trait anxiety might differentially affect variables that have been commonly associated with state anxiety arousal. Some researchers have suggested that state anxiety reactions may differ in quality according to the situations that arouse them.
Thirty-two subjects were placed in four groups according to their scores on the social evaluation (SE) and physical danger (PD) subscales of the S-R GTA. A high general anxiety group was comprised of volunteers who scored high on both subscales. A low general anxiety group was comprised of volunteers who scored low on both subscales. Two groups reporting moderate levels of general anxiety were comprised of volunteers who scored low on one subscale while scoring high on the other. The subjects were exposed to two stressing situations, one involving social evaluation (a purported test of intelligence) and another which involved physical danger (threat of electric shock). The experimental design was a 4 (groups) by 2 (situations) latin square with repeated measures. Measures of self-reported state anxiety, heart rate, skin conductance, self-reported physiological arousal, performance on a simple task, self-efficacy and self-evaluation were obtained for each situation.
The increments in self-reported state anxiety to the stressing conditions paralleled the expectations based on the respective subscale scores on the S-R GTA. Subjects who reported similar levels of trait anxiety to either dimension, regardless of their levels of general anxiety, did not differ in their respective increases in state anxiety to stress. In a discriminant analysis of responses in all of the dependent variables to social evaluation stress and physical danger stress, 46 of the 64 observations were correctly classified. Subjects reported lower levels of self-efficacy to the task involving physical danger, but evaluated their performance as much better than on the task involving social evaluation. The subjects also self-reported higher levels of physiological arousal to the physical danger task, but did not experience greater increases in heart rate or skin conductance.
The results provide additional support for the multidimensional view of trait anxiety and argue against the use of a total score on the S-R GTA as a measure of general trait anxiety. The total score did not enhance the ability of the subscale scores to predict state anxiety responses to trait-congruent situations. The results also provide support for the view that the dimensions of trait anxiety differentially arouse the various components of state anxiety. Theoretical and practical implications of the results were discussed and suggestions for future inquiry were offered.
© The Author
Is Part Of
VCU University Archives
Is Part Of
VCU Theses and Dissertations
Date of Submission