Previously existing computer analysis programs have been unable to correctly identify many complex syntactic structures thus requiring further manual analysis by the clinician. Complex structures, including the relative clause, are of interest in child language samples due to the difference in development between children with and without language impairment. The purpose of this study was to assess the comparability of results from a new automated program, Cx, to results from manual identification of relative clauses. On language samples from 10 children with language impairment (LI), 10 language matched peers (LA), and 10 chronologically age matched peers (CA), a computerized analysis based on probabilities of sequences of grammatical markers agreed with a manual analysis with a Kappa of 0.88.
College and Department
David O. McKay School of Education; Communication Disorders
BYU ScholarsArchive Citation
Michaelis, Hali Anne, "Automated Identification of Relative Clauses in Child Language Samples" (2009). All Theses and Dissertations. 1997.
relative clause, computer analysis