Measurement of Defenses in Special Populations: Revision of the Defense Mechanisms Inventory

The Defense Mechanisms Inventory (DMI) has been standardized for the general and minority populations and used in counseling settings and in scores of interpersonal psychology studies. Guidance and research experience with the DMI has indicated that the requirement of choosing most representative and least representative item options can introduce inaccuracies and promote peculiar response sets, particularly among certain respondents with learning disabilities and those with lower reading skills or poor educational backgrounds. The DMI protocols of 504 undergraduate college students and 229 minority youngsters were rescored, utilizing only the most representative responses. In addition, data were collected from 32 graduate students using modified instructions (asking for most representative choices only), as well as standard instructions. Data were also analyzed from 127 male drug detoxification patients who completed the DMI with modified instructions. The reliability and factor structure of the defensive and object-relations scores were compared to standardization data for the traditional scoring method. Results point to the merits of using only the most representative instruction with special populations.