The purpose of this study was to examine the effects of integrated art therapy upon self-efficacy of broken family child, and also investigate its effects upon the subcategory of self-efficacy such as confidence, self-control efficacy and task-difficulty preference. The subjects of this study were 10 children selected from broken family children requested by a center director at a local children center located in I city in Jellabuk-do, and they were divided into two groups, an experimental group of five children and a control group of five children. A pre-test was conducted for the two groups respectively using self-efficacy scale, in 16 sessions held for 45 minutes per session in a period between February 27 and April 20, 2006. The program that the center has in use was applied to the control group, and an integrated art therapy for self-efficacy improvement performed for the experiment group. Following the completion of all sessions, a post-test was conducted in the same manner as the pre-test, and like pre-test condition, the data of 10 children were analyzed with t-test. The results showed no significance of self-efficacy between pre- and post- test for both control and experimental group, t-test was employed to analyze the significance in the points of experimental group resulted from confidence, self-control efficacy and task-difficulty preference, which are a sub-category of self-efficacy. From the results, no statistical significance was shown in the pre- and post-test between the one the art therapy was applied to and the other the art therapy was not applied to. However, the group the art therapy was applied to, was showed to be a statistically significant difference in self-efficacy improvement, and the confidence out of confidence, self-control efficacy and task-difficulty preference that are the sub-category of self-efficacy, only showed a statistically significant difference. And it was assumed that there was a positive change in the fields of self-control efficacy and task-difficulty preference since their average values showed high in pre- and post-test. However, in the case of control group that no treatment was applied to, the average values in pre- and post-test were decreased, and showed no significant difference. This suggests that the integrated art therapy may have generally enticed a positive change in self-efficacy improvement of broken family child.