utility variables were first conducted using a 3 (case) x 2 (model) mixed MANOVA with the two models treated as repeated measures. Significant main effects were found for model [F(6, 235) = 57.13, p < .001], and for case [F(12, 470) = 1.78, p = .049]. There was no interaction between case and model [F(12, 470) = 1.16, p = .31]. Post hoc tests were conducted using a Bonferroni correction.
For Ted, there were no significant differences between the models with respect to the clinicians’ ratings of the ease of application or the usefulness for professional communication. However, as shown in Table 3, there was a significant difference in favor of the five-factor model on the remaining four aspects of clinical utility. The clinicians rated the FFM as more useful for communicating information to clients or other laypersons, F(1, 72) = 57.03, p < .001. The FFM was also rated significantly higher than the DSM in terms of the model’s ability to adequately describe all of the individual’s important personality difficulties, F(1, 72) = 24.21, p < .001, and for providing a global personality description, F(1, 72) = 8.42, p = .005. Finally, the clinicians’ mean ratings were also greater for the FFM with respect to the utility of the model for making treatment decisions, F(1, 72) = 8.10, p = .006.
Clinicians’ ratings of clinical utility followed a similar pattern for Earnst, with one exception. There was again no significant difference between the DSM and FFM ratings in terms of ease of application. The clinicians again rated the FFM as having significantly greater utility than the DSM in terms of communication with their clients, comprehensive description of the individual’s important personality difficulties, treatment planning, and global personality description (see Table 3). The one deviation from the findings obtained for Ted was that the