Automated Identification of Adverbial Clauses in Child Language Samples

• Main Author: Clark, Jessica Celeste
• Format: Others
• Published: BYU ScholarsArchive 2009
• Subjects: automated analysis; adverb clause; adverbial clause; complex syntax; Cx; intervention; language analysis; language impairment; language sample; manual analysis; Communication Sciences and Disorders
• Online Access: https://scholarsarchive.byu.edu/etd/1682; https://scholarsarchive.byu.edu/cgi/viewcontent.cgi?article=2681&context=etd

In recent years, computer software has been used to assist in the analysis of clinical language samples. However, this software has been unable to accurately identify complex syntactic structures such as adverbial clauses. Complex structures, including the adverbial clause, are of interest in child language due to differences in the development of this structure between children with and without language impairment. The present study investigated the accuracy of new software, called Cx, in identifying adverbial clauses. Two separate collections of language samples were used. One collection included 10 children with language impairment, 10 age-matched peers, and 10 language-matched peers. A second collection contained language from 174 students in first grade, third grade, fifth grade, and junior college. There was high total agreement between computerized and manual analysis with an overall Kappa level of .895.