care for school-age children in hospitals. The present study was carried out in an attempt to obtain data on this aspect of the question. Previous studies of the attitudes of adults toward children's behavior problems have indicated that persons with little or no knowledge of mental hygiene tend to rank as fairly or very serious those problems that involve actions conflicting with the social code, and that they underrate the significance of withdrawing tendencies, recessive personality, and behavior traits, which mental hygienists regard as most serious. Stogdill (5) compared the ratings given various behavior problems by parents and by mental hygienists. The ratings of his mental hygienists agreed for the most part with those of the Wickman study; while parents, like teachers, were inclined to rate overt behavior problems as the more serious. Bain (2) found that Teachers College students come nearer to the mental hygiene point of view on problem behavior after they had had one or more courses in child study. Preston and Shapler (4) comparing a group of supposedly normal grade school children with a group classed as needing psychiatric treatment found that the control group could not be distinguished from the clinic group on the basis of behavior traits. Temper tantrums, for instance, occurred in an equal percentage of both groups, while daydreaming was more prevalent in the control 'group than in the clinic group. This indicates the wide variance in opinion as to what different people consider as McFie (3) concluded from a study of children referred to a psychiatrist that parents tend to regard active disturbing behavior as much more serious than the personality deviations that give concern to the psychiatrist. Only one recent investigation, that of Arlitt and Lloyd (1), has involved a study of the problem behavior exhibited by children in hospitals. These authors, however, studied preschool children, and their aim was to discover the effect of hospitalization as a factor in producing behavior problems. The object of the present study was to learn the attitudes of nurses toward various types of problem behavior, rather than to search into possible causes of the behavior. Future studies may be able to combine the two points of view and thus add to our knowledge of the cumulative and interactive effects of behavior. Subjects and method.--It was not possible to find a group of nurses comparable in both education and term of service to answer the questions used in this study, under even partially controlled conditions. The subjects used were student nurses who had had some experience in pediatric nursing and graduate nurses actively engaged in nursing children. Some of both groups had had only high school education before their hospital training, while others had had a year or more of college work. Many of the college girls had completed a four-year course. Some