UBC Theses and Dissertations
The reliability and internal consistency of the thematic apperception test Epstein, Marilyn
The purpose of this study was to investigate the repeat reliability and internal consistency under short-term conditions of several indices of agression and anxiety as measured by the TAT. In view of the variations in the results reported in the few studies concerned with this problem, a specificity hypothesis was suggested. This hypothesis states that no general evaluation can be made of the temporal and internal stability of the TAT. Such statements probably only have meaning in terms of specific variables. The variables employed in the present study were aggression and anxiety and the results should not be generalized beyond these variables. One group of subjects was given standard TAT instructions at two successive administrations, while a second group was asked to tell a different story to each card. This procedure was designed to control and study the influence of memory effects. It was found that memory effects are very strong, and where the instructions interfere with their operation, repeat reliability coefficients are very low. The TAT cards included two high, two medium and two low aggressive content cards, as determined by a panel of judges and from previous research. The purpose of this part of the study was to determine if the reliability of the test varies with the level of card ambiguity for a given drive. The results did not support the hypothesis that responses to stimuli which are unambiguous for a given drive are more likely to be stable over time than responses made to a relatively ambiguous stimulus. The internal consistency was evaluated by correlating the scores obtained on the first session by all subjects in terms of the level of ambiguity. These correlations were quite low, indicating the need for caution in using an additive treatment of scores from different TAT cards.
Item Citations and Data