Abstract

Language samples are naturalistic sources of information that supersede many of the limitations found in standardized test administration. Although language samples have clinical utility, they are often time intensive. Despite the usefulness of language samples in evaluation and treatment, clinicians may not perform language sample analyses due to the necessary time commitment. Researchers have developed language sample analysis software that automates this process. Coding schemes such as that used by the Systematic Analysis of Language Transcripts (SALT) software were developed to provide more information regarding appropriate grammatical tag selection. The usefulness of SALT precoding in aiding automated grammatical tagging accuracy was evaluated in this study. Results indicate consistent, overall improvement over an earlier version of the software at the tag level. The software was adept at coding samples from both developmentally normal and language impaired children. No significant differences between tagging accuracy of SALT coded versus non-SALT coded samples were found. As the accuracy of automated tagging software advances, the clinical usefulness of automated grammatical analyses improves, and thus the benefits of time savings may be realized.

Degree

MS

College and Department

David O. McKay School of Education; Communication Disorders

Rights

http://lib.byu.edu/about/copyright/

Date Submitted

2015-06-01

Document Type

Thesis

Handle

http://hdl.lib.byu.edu/1877/etd7794

Keywords

language sample, language impairment, automated tagging, software, SALT

Share

COinS