Inpatient aggression has been increasingly problematic in psychiatric facilities across the United States and around the world. Psychological assessment measures, such as the Rorschach Inkblot Method, are often used in psychiatric facilities to clarify a patient's diagnostic issues and assist in treatment planning. An assessment measure that could provide information about the type, intensity, and direction of a patient's aggressive impulses would therefore be clinically useful. The current method for scoring aggression on the Rorschach provides limited information about a patient's aggressive drives; Gacono & Meloy have proposed a broader system for scoring Rorschach aggressive content. Thus far, research on this new aggression scoring system has neglected to examine patients with Axis I disorders. The purpose of this dissertation was to explore the differences between the types and frequencies of these newer aggression variables, as well as the utility of these scores in predicting aggression in an inpatient sample of depressed and paranoid patients. This sample was chosen based on psychoanalytic conceptualization of aggressive dynamics in these patients. Results of Poisson and negative binomial regressions indicated that there were no differences between the depressed and paranoid groups in terms of the types or frequencies of Rorschach aggressive content. Kruskal-Wallis tests indicated that there were some differences between the groups in terms of the type and severity of behavioral manifestations of aggression. Finally, a logistic binomial regression showed that Rorschach variables did not add significantly to the prediction of the presence of aggressive behavior in this population. Clinical implications, limitations of the study, and directions for future research are examined. / text