A Normative Study of TAT Ambiguity

Type D personality refers to a clustering of 2 stable personality traits, namely negative affectivity and social inhibition. Currently Type D is standardly assessed using the DS14. An experimental Type D personality scale, the DS⁽³⁾, was developed to examine an avenue for assessing Type D more efficiently. The DS⁽³⁾ differs from the DS14 in its use of a 3-point Likert scale to rate responses, use of all negatively worded items, and a rearranged presentation of items. This article examines the psychometric properties of this questionnaire by examining its dimensionality, item and scale properties, and cutoff scores to screen for Type D personality. Data from 2 clinical samples were analyzed using item response theory. The results suggest that the DS⁽³⁾ is a potentially suitable instrument for Type D assessment. It has high reliability, and Type D personality classification based on this scale corresponds well with the current standard Type D assessment based on the DS14.     

Exploring fantasies of TAT reliability

It has been argued that low TAT stability estimates are an artifact of retest instructions to be original and creative. This view was supported in a recent study by Winter and Stewart (1977) who found that test-retest reliability was higher when subjects were told on the second occasion to write the same stories than when they were told to write different stories. This study attempted to replicate Winter and Stewart's results and also to determine whether the similarity of stories did vary as hypothesized. The replication was only partially successful. More importantly, it was concluded that while instructions to write similar stories did improve reliability, the increase reflected only subjects' memory capabilities and not greater stability in motive expression. While true TAT stability may still be commonly underestimated, altering retest instructions to improve reliability does not appear to be the solution.     

A Scoring Scheme for the TAT and other Verbal Projective Techniques

Summaiy: Three groups of subjects rated each item in Rotter's I-E Scale for its social desirability value. Social desirability scale values (SDSV) of paired items were compared with one another. Results indicated that paired items are not sufficiently similar in their SDSVs, thus enabling subjects to respond on the basis of social desirability. All three groups rated internal items as being significantly more socially desirable than external items. The correlations between the SDSV ratings of the three groups were highly significant. The results indicate that the scale, as it is currently used, is seriously deficient.     

Treatment of Time in the TAT

Part of a larger empirical investigation to identify and classify subjective probability notions and explore their relationship to personality variables (Lipgar, 1965). Four factors that distinguished four types of subjective probability notions had been obtained from a matrix of intercorrelations among 30 Ss' Q sort ratings on patterns of “head-tail” guesses. To investigate the possibility that subjective probability notions are associated with other orientational activities in situations of uncertainty, a typology of time orientations and rating scales was derived from TAT stories by those Ss who best represented the four subjective probability factors. These and other stories were rated by four judges independently, and an association between treatment of time in the TAT and subjective probability notions was substantiated.     

A measure of needs to assess the stimulus characteristics of TAT cards

Summary: The stimulus demands of TAT cards were evaluated by means of a TAT-Adjective Rating Scale. Seventeen needs were measured through adjective ratings. Mean needs were reported for 191 males and 85 females. Variability was used to develop a measure of stimulus ambiguity. Also investigated was (a) the partitioning of variance for subjects, cards, needs and their interactions, and (b) sex differences were evaluated by means of analysis of variance.     

A Note on the Stability of TAT Interjudge and Intrajudge Ambiguity Scores

The stability of TAT ambiguity values, based either on interjudge, or intrajudge response variability, was investigated. Interjudge variability was estimated from dispersion of the first response emitted to each stimulus; intrajudge variability utilized multiple associations by Ss. Multiple association yielded highly stable values in comparisons over time, and neither method resulted in stability across S samples. It was concluded that the definition of ambiguity must include intrajudge variability in order to achieve a reliable ordering of stimuli for a given respondent.