A feasibility study of the use of language transcription and analysis software by non-SLTs

TitleA feasibility study of the use of language transcription and analysis software by non-SLTs
Publication TypeConference Paper
Year of Publication2011
AuthorsWren, Y, Overton, S
Conference NameChild Language Seminar
Date Published06/2011
Abstract

Background. Language sample analysis is widely recognised as an important part of the assessment process. However, speech and language therapists (SLTs) are often unable to dedicate the time necessary to transcribe and formally analyse a language sample. Instead SLTs may make auditory judgments of a child’s language production to guide their planning of intervention but use standardized assessments for target setting and outcome measurement. For some children however, such assessments are unsuitable for identifying appropriate targets or showing outcomes and a more functional yet objective measure, such as language sample analysis, is needed.
Aims. This study aimed to ascertain the reliability of non-SLTs’ transcriptions of child language samples using language analysis software as a way of circumventing the problem of time needed for language transcription. A secondary aim was to propose a feasible way of bringing standardised language sampling methods into services and clinical practice.
Methods & Procedures. A Speech and Language Therapy Assistant (SLTA) was trained to transcribe child language samples using Systematic Analysis of Language Transcripts (SALT) software. Fifteen narrative language samples were audio recorded by Speech and Language Therapists working within the NHS. The samples were transcribed by the SLTA and the SLT independently and inter-rater reliability calculated. Analyses were then returned to SLTs for interpretation and comments about the usefulness of the data and the practicalities of the process.
Results and outcomes. Non-SLTs can carry out reliable language transcription using SALT. SLTs reported that the individualised nature of the data produced was useful for target setting but that data presentation could be more clinically tailored.
Conclusions & Implications. Employing non-SLTs to transcribe samples would be a cost-effective way of implementing rigorous, standardized and highly individual language assessment within over-stretched services. A further study exploring the potential of computerised analysis for objective outcome measurement is in progress.