교사보고형 아동 언어능력 평가척도(TROLL) 개발을 위한 예비연구

Objectives: When a preschool child is suspected or having a delay in language development, an early intervention via an expert is needed. If teachers, who are capable of observing a large number of children in the course of education for a large-scale peer group, could screen children with suspected language delays, it will increase access to further early language interventions. Thus, this study set out to validate a teacher-reported preliminary inventory for children's language abilities. Methods: The inventory used in this study was based on Dickinson, McCabe, & Sparague (2003) and was revised and supplemented items to measure three subcomponents (language use, reading, and writing) of children's language abilities. The study included 74 4- to 6-year-olds and six teachers. Item quality, inter-item consistency, and construct validity were analyzed. For validity testing related to criteria, two speech-language pathologists conducted standardized tests on vocabulary and language and reviewed their correlations with the inventory. Results: The test results of item quality show that the correlation coefficient between sub-constructs (reading and writing) was .849, which led to the combination of the two sub-constructs into one (early literacy). Other analyses generated all proper results, but the criteria validity results show no significant correlations between the sub-constructs of language use and official tests. Conclusion: The teacher-reported inventory for children's language abilities was found to be a good test to measure preschool children's language abilities in Korea. Recognizing a need to revise the content of items of lower discrimination, the study made proposals for their revision.