Behavioral correlates of psychometric patterns : A factorial study

The present Investigation was an attempt to demonstrate a relationship between particular psychometric patterns and external criteria. The hypothesis of the study was that a relationship exists between aspects of psychometric performance and other measures of behavior, and that this relationship could be shown by means of a correlational analysis. The subjects for this study were a group of randomly selected female psychiatric patients at the University of Texas Medical Branch Hospitals. The subjects included most of the usually diagnosed neuroses and psychoses with the exception of neurological disorders. Behavioral items utilized in this study were gathered from the ease histories of the subjects and, in general, were behaviors frequently associated with clinical concepts of psychopath, delinquency, character disorder, and hysteria. The test variables used were certain Wechsler-Bellevue test scores as well as certain qualitative aspects of the content of some Rorschach test responses. The hospital records of all subjects were checked for the presence or absence of the behavioral and psychometric items and the data intercorrelated by means of tetrachoric correlations. The correlation matrix was factor analyzed by the complete centroid method and the eight resulting factors rotated orthogonally to simple structure. The following factors were extracted: Factor I - Inadequate Socialization. The behavioral items loading on this factor were identified as a constellation of behaviors in which the common characteristic is- an inadequate assimilation of the values and restrictions of the culture and a rebellion against them. The test items of high Performance IQ, number shock, and low Arithmetic significantly loaded-this factor and were considered as fairly pure measures of it. It was suggested that this factor is similar in context to Cattell's Factor C. Factor II - Surgency. The second factor was identified as a constellation of behaviors in which the major characteristic was surgency. The test items loading on this factor were low Information to Comprehension and Picture Completion plus Picture Arrangement greater than Block Design plus Digit Symbol. This factor seemed to be- a good match with Cattell's Factor F. Factor III - Hypochondraisis. The third factor was identified as a hypochondriacal factor. The test item loading on this factor was Immature M, and it was suggested that immature content on the Rorschach is related to similar attitudes and expressions of immaturity. Factor IV - Hysteria. The behaviors loading on this factor were considered to be similar to those frequently associated with the clinical concept of hysteria. Aside from low intelligence, no test item loaded this factor. A tentative match was made between it and Cattell's Factor I. Factor V - Immature Dependency. This factor was identified as a constellation of behaviors in which the common unity is the expression of immaturity and dependency. Immature M significantly loaded on this factor, and it was concluded that the factor was similar in context to Cattell's Factor G. Factor VI - Aggression. This factor was significantly loaded with items identified as having a common unity of hostility or aggression. The presence of aggressive movement responses on this factor was taken as evidence for the relationship between this qualitative aspect of the Rorschach test and other indications of aggressive or hostile behavior. Factor VII - Anxiety. This factor was tentatively identified as an anxiety factor. Anatomy and sex responses on the Rorschach significantly loaded the factor, and it was concluded that there was a relationship between these responses and other expressions of anxiety. Factor VIII (residual factor). This factor was considered to be a residual factor and no attempt was made to interpret it. With one exception (Rorschach color responses), all test items significantly loaded on one or more of the seven significant factors. Loadings of .50 to .89 could occur by chance only very infrequently, and therefore they were taken to indicate a strong relationship between the test items and the personality variables represented by the factor space. Each of the seven factors loaded both behavioral and psychometric data. This demonstration of common factor variance was considered to be good evidence for support of the hypothesis that a relationship exists between psychometric and behavioral data.