Transcription of language samples is labor intensive and data on intra- and inter-annotator reliability is lacking. Mean Length of Utterance in Morphemes (MLUM) can be calculated from transcripts and used to evaluate intra-annotator reliability. We developed a computational method to determine transcriber consistency across a large speech corpus.
Details
Title
Using MLU to evaluate the reliability of ADOS transcription
Creator
MacFarlane, Heather : Oregon Health and Science University Salem, Alex : Oregon Health and Science University Dolata, Jill : Oregon Health and Science University Lawley, Grace : Oregon Health and Science University Fombonne, Eric : Oregon Health and Science University
Meeting Name
Research Week, Oregon Health and Science University, 2020