아동용 교우관계문제검사의 개발과 타당화

Development and Validation of Inventory of Peer Relation Problems for Elementary School Children

This study has been carried out with the aims of developing a comprehensive inventory of peer relation problems, which is based on the Korean Inventory of Interpersonal Problems Circumplex scales (KIIP-C). It also aims to examine reliability and validity of the inventory, and provide a preliminary norms. For the current study, inventory items were culled from the following sources: the Korean Inventory of Interpersonal Problems Circumplex scales (KIIP-C), various current scales dealing with peer relations of children, and a survey of children's peer relation problems. The preliminary items for each scale were administered to 220 fourth through sixth graders. The resulting skewness of distribution, kurtosis, mean and standard deviation, item-total correlation, internal consistency, and meanings of the items were comprehensively considered in selecting the final 64 items. In order to check on reliability, internal consistency, convergence and discrimination reliability of the final items and scales, the data were collected from 1,046 fourth through sixth graders currently attending four elementary schools. The study results can be summarized as follows. Internal consistency of the inventory of peer relation problems showed the range between .70-.94 (median value of .75), split-half reliability between .67-.83 (median value of .75), and test-retest reliability between .69-.88 (median value of .81). Inter-correlation of 8 scale scores and factor analysis results of individual ipsative scores showed that the circumplex property of inventory of peer relation problems is appropriate. Regarding correlations between various existing indices and scales related to peer relation problems, both convergence reliability and discrimination reliability were found to be fair. When the scale scores for the inventory of peer relation problems compared according to the factors of gender and grade, the primary effects of gender and grade were statistically meaningful whereas effects of interaction between gender and grade were not. This study can be considered meaningful in that it constructed an inventory for a comprehensive evaluation of peer relation problems specific for children and provided preliminary norms.