Psychological Assessment 1992, Vol. 4, No. 2,156-159 Copyright 1992 by the American Psychological Association, Inc. 1040-3590/92/S3.00 Performance of Disruptive Behavior Disordered and Normal Samples on the Draw A Person: Screening Procedure for Emotional Disturbance Jack A. Naglieri Department of Educational Services and Research The Ohio State University and The Institute of Clinical Training and Research, The Devereux Foundation Devon, Pennsylvania Steven I. Pfeiffer The Institute of Clinical Training and Research, The Devereux Foundation Devon, Pennsylvania Draw A Person: Screening Procedure for Emotional Disturbance (DAP:SPED) scores were com- pared for 54 normal students and 54 students with conduct and oppositional defiant disorders who attended a psychiatric day treatment facility. Both samples were age 7-17 years. The samples were also matched by sex, race (White/Non-White), and geographic region. The DAP:SPED mean T score earned by the 54 subjects in the clinical sample (M= 56.63, SD = 10.27) was significantly higher (t = 4.05, p < .001) than that of the 54 normal subjects (M= 49.37, SD = 8.68), indicating that the clinical group produced more signs associated with emotional disturbance than did the normal group. Further analysis indicated that use of the DAP:SPED increased diagnostic accuracy by 25.8%. These results suggest that this new approach to scoring the man, woman, and self drawings may be more useful than previous human figure drawing techniques. Human figure drawings, such as the House-Tree-Person, Ki- netic Family Drawings, and Draw-A-Person have been used very frequently by psychologists in clinical practice (Cum- mings, 1986; Lubin, Larsen, & Matarazzo, 1984). The Draw-A- Person has been especially useful because it can be used to evaluate both the emotional (Koppitz, 1968; Machover, 1949) and intellectual (Harris, 1963; Naglieri, 1988) status of children and adolescents. Among the most well-known approaches to the use of human figure drawings to obtain information about emotional status were those developed by Buck (1948) and Ma- chover (1949). Of the two, Machover's text, Personality Projec- tion in the Drawing of the Human Figure, has had the most influence on how children's drawings are interpreted (Cum- mings, 1986). This approach involves interpretations of individ- ual signs appearing in the drawings of human figures, such as the omission of eyes, as being associated with a particular inter- pretation (e.g., unwillingness to interact with the environment). Although Machover's interpretation system and others like it, such as Kinetic Family Drawings, have been very widely used, researchers have not found much support for the association of an individual sign with a specific meaning (Roback, 1968; Swensen, 1957,1968). An alternative to interpretation of individual signs has been proposed that examines the drawings of human figures from a We thank the staff and clients of the Devereux Foundation Day School for their cooperation, the research assistants from the Institute of Clinical Training and Research for the data collection, and Fran Mues for her assistance in editing an earlier draft of this article. Correspondence concerning this article should be addressed to Jack A. Naglieri, Department of Educational Services and Research, The Ohio State University, 356 Arps Hall, 1945 North High Street, Colum- bus, Ohio 93210-1172 holistic or global perspective (Koppitz, 1984). This approach involves examining the number of times signs considered to be associated with disturbance are produced in comparison with the number of times these same signs are found in the drawings of normal persons. This is more consistent with current views that the Draw-A-Person should be used as a global indicator of emotional status rather than to develop specific psychological interpretations (Tharinger & Stark, 1990). Increased reliability and validity evidence, as well as good standardization samples, are needed for the Draw-A-Person technique to meet current standards (e.g., American Education Research Association, American Psychological Association and National Council on Measurement in Education, 1985) necessary for any test. Even though past projective tests have not had adequate psychometric qualities, (e.g., Anastasi, 1988, reported that projective tests typically have reliabilities of less than .20), attempts to rectify this problem for the Draw-A-Per- son have been made by Koppitz (1984). More recently, Naglieri, McNeish, and Bardos (1991) also provided evidence of in- creased reliability, validity, and national norms. Koppitz's (1984) approach provided a transition from past interpretation schemes to one that is based on the following three criteria: (a) items should differentiate between healthy and emotionally disturbed children; (b) the frequency of occur- rence of signs should be low (<16%) in the normal population; and (c) ratesof occurrence should be independent of age. Follow- ing these guidelines, Koppitz identified a group of 30 items to identify subjects with emotional disturbance. Her approach, however, is not standardized, its internal reliability has not been adequately examined, and the evidence of its effectiveness is limited. Moreover, in a recent investigation, Tharinger and Stark (1990) found that neither a total score nor any of Koppitz's 30 emotional indicators differentiated children with emotional disorders from a normal sample. 156 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.