Journal of Counseling 1972, Vol. 19, No. 2, 94-103 A MULTIMETHOD FACTOR ANALYSIS OF THE DIFFERENTIAL PERSONALITY INVENTORY, PERSONALITY RESEARCH FORM, AND MINNESOTA MULTIPHASIC PERSONALITY INVENTORY D. MERILEE TROTT ANDMARTIN E. MORI* University oj Windsor, Ontario, Canada The Differential Personality Inventory, Minnesota Multiphasic Personality Inventory, and Personality Research Form were administered to 151 college students seeking psychological services. The scale scores were subjected to a multimethod factor analysis. The first 19 factors were rotated to a Varimax criterion and interpreted. The rotated factor matrix indicates that a con- siderable proportion of the variance of both the Differential Personality Inventory and MMPI scales is attributable to the same factors, that the Differential Personality Inventory scales are factorially less complex than the MMPI scales, and that psychopathological behavior is intimately linked with interpersonal behavior in general and is best regarded as quantitatively rather than qualitatively different from normal behavior. Categories of classification, diagnostic la- bels, and constructs used in psychopathology have caused considerable dissatisfaction. Zigler and Phillips (1961) have criticized the lack of homogeneity, reliability, and validity of diagnostic categories and have recommended a correlational and descrip- tive approach as opposed to a theoretical and etiological or explanatory approach. Szasz (1957) suggested that different noso- logical systems, based, for example, on either a physiological, mental, or behavioral frame of reference, be used in different con- texts. Perhaps the most fruitful attempt to deal with this classification problem was Eysenck's (1961) factor analytical ap- proach. His results demonstrated that it is helpful to think of subjects located in an Tir-dimensional space rather than in either Category A or B. Thus, for example, a per- 1 Requests for reprints should be sent to D. Merilee Trott, Department of Psychology, Univer- sity of Windsor, Windsor, Ontario, Canada. "Grateful acknowledgment is extended to Douglas N. Jackson for helpful suggestions and to the faculty of the Psychological Services Center of the University of Windsor for their assistance in the collection of the data. son is not neurotic or psychotic but rather is located on each of the factor analytically established dimensions of neuroticism and psychoticism. Some research on the Minnesota Multi- phasic Personality Inventory (MMPI) is directly relevant to this controversy. The inventory was devised to differentiate se- lected psychiatric criterion groups from nor- mals (Hathaway & McKinley, 1940). Stud- ies involving judgmental (Harris & Lin- koes, 1955) and factor analytical (e.g., Comrey, 1957; 1958) methods have demon- strated, however, that the MMPI scales do not reflect clearly unitary constructs. The items comprising the standard MMPI scales are relatively heterogeneous and consist in turn of more homogeneous subsets or clus- ters of items. Questions concerning the constructs best used in the explication of psychopathologi- cal behavior thus remain to be answered and Eysenck's work suggests that factor ana- lytical approaches may be fruitful. A factor analytical approach requires an initial bat- tery of scales or items measuring the domain to be structured. The more adequate the 94