Defense Date


Document Type


Degree Name

Doctor of Philosophy



First Advisor

Edwin R. Thomas


Thirty six female intercollegiate team athletes and 40 female non-athlete control subjects were studied in a two-experiment investigation designed to explore the personality patterns of female athletes; to measure any differences in performances of the experimental and control groups attributable to changing conditions, i.e., solo, coaction and competitive; and to explore the interactions of personality variables and performance. The previous research on personality, with Cattell's Sixteen Personality Factor Inventory and female athletes, is not plentiful and frequently in conflict. No previous research could be located that employed controlled competitive conditions with athletes—male or female. In Study 1, the team athletes were found to be more tough-minded (1-) and more group dependent (Q2-) than controls. The athletes were also found to be more tough-minded, group dependent, assertive (E+), venturesome (E+), and practical (M-) than Cattell's female college normative group. In a post-hoc analysis, the controls were found to be different from test norms on four scales. Discussions of the sometimes conflicting results is offered. A post-hoc discriminant analysis was also performed and discussed. The factor scales which were found to discriminate, in order of appearance, were Q2, A, I, B, M, and Q4-

In Study 2, all subjects were administered one of the three experimental conditions on a stationary bicycle-competition, coaction (performance with another), and solo performance. An ANOVA applied to the 2 x 3 design found the conditions to be significant but not the Subject groups or inter-actions of subject groups and conditions. A sub-analysis showed the competition condition to be significantly different from each of the other two conditions.

To investigate the interactions between personality and performance, a regression analysis was performed to test which of the 16PF scales best predicted competitive performance. Ten variables accounted for 23,81 of the variance at the .01 level of confidence. The variables, in order of appearance, were factors I, Q1, A, H, B, C, Q2, E, Q3, and M.

Since the use of t-tests with groups of athletes has proven fruitless, it is recommended that future research in the area of sports personality use a different methodological approach. The only exception would be if the groups of athletes are highly unique (for example, fencers only) and at the highest levels of successful competition. The discriminative analysis procedure appears to hold some promise as does the application of a complex motor task under controlled conditions. Finally, the study of interactions of personality variables with varying conditions of performance seems to offer a promising area for further investigations.


Scanned, with permission from the author, from the original print version, which resides in University Archives.


© The Author

Is Part Of

VCU University Archives

Is Part Of

VCU Theses and Dissertations

Date of Submission


Included in

Psychology Commons