PROBLEM has been used extensively to study numerous correlates of childhood anxiety, e.g.,' intelligence, achievement, sociometric status, and behavior with Lipsitt (g) reported a negative correlation between children's anxiety, as measured on the CMAS anxiety scale (A scale), and self-concept. This finding received corroboration from a number of subsequent investigations(2-4 , 12). Examination of the literature, however, uncovered no studies of relations between A scale performance and self-concepts of emotionally disturbed children in residential treatment. In the present study, it was predicted that a negative relationship between A scale scores and self-concept would obtain for disturbed, as well as normal, children.In addition to the A scale, the CMAS contains a lie scale (L scale), devised to measure the tendency to present oneself in an extremely favorable light. It is surprising that past investigations of the A scale and self-concept have not studied relationships between the L scale and self-concept. If the L scale measures a selfaggrandizement tendency, it seems reasonable to expect a direct association between the L scale and self-concept, i.e., the higher a person scores on the L scale, the more positive will appear his self-concept. This hypothesized relationship was examined in the present study.13, 1 1 ) on the CMAS have reported either a zero-order or low negative correlation between the A and L scales. A further purpose of this study was to examine the relationship between the two scales in a group of disturbed boys. Several studies@; 6 -8, 1') comparing normal children with non-institutionalized emotionally disturbed children have shown higher A scale scores for the disturbed children. The institutionalized boys in the present study were, therefore, expected to score higher on anxiety than did Lipsitt'scg) normal boys.The Children's Manifest Anxiety Scale (CMAS)
13).Normative studies", 6 -METHOD Subjects. All testable boys (N = 44) in residential treatment a t the Emma Pendleton Bradley Hospital participated in this experiment. They ranged in age from 7 to 14 years, mean 10.8 years. The mean length of time in residential treatment was 23 months.Procedure. Each S was tested individually by one of three female experimenters (Es). They were asked to provide social evaluations for a list of words (social evaluation task), then to complete the CMAS, and finally to describe themselves by use of the same word list (self-description task).Word list. The list consisted of 66 adjectives describing a wide range of traits, 36 of positive value and 30 of negative value, as determined by two independent judges. Social evaluation task. E read the following instructions .to S : "I want to ask you some questions about how you feel about other boys. You just answer 'yes' or 'We gratefully acknowledge the assistance of Linda Pappas, Edith Leverens, and Sandy Weir who participated in this research as undergraduates in an experimental child sychology course at Brown University.
