Department of Communication Sciences and Disorders, Northeastern University, Boston, MA.
Department of Communication Sciences and Disorders, University of Wisconsin-Madison.
Am J Speech Lang Pathol. 2021 Jun 18;30(3S):1558-1571. doi: 10.1044/2020_AJSLP-20-00144. Epub 2021 Mar 1.
Purpose The objectives of this study were to: (a) compare interrater reliability of practicing speech-language pathologists' (SLPs) perceptual judgments of phonetic accuracy and hypernasality between children with dysarthria and those with typical development, and (b) to identify speech factors that influence reliability of these perceptual judgments for children with dysarthria. Method Ten SLPs provided ratings of speech samples from twenty 5-year-old children with dysarthria and twenty 5-year-old children with typical development on two tasks via a web-based platform: a hypernasality judgment task and a phonetic accuracy judgment task. Interrater reliability of SLPs' ratings on both tasks was compared between children with dysarthria and children with typical development. For children with dysarthria, four acoustic speech measures, intelligibility, and a measure of phonetic accuracy (percent stops correct) were examined as predictors of reliability of SLPs' perceptual judgments. Results Reliability of SLPs' phonetic accuracy judgments and hypernasality ratings was significantly lower for children with dysarthria than for children with typical development. Among children with dysarthria, interrater reliability of perceptual judgments ranged from strong to weak. Percent stops correct was the strongest predictor of interrater reliability for both phonetic accuracy judgments and hypernasality ratings. Conclusions Reliability of perceptual phonetic accuracy judgments and hypernasality ratings among practicing SLPs for children with dysarthria is reduced compared to ratings for children with typical development. Findings underscore the need for more reliable methods to assess phonetic accuracy and hypernasality for children with dysarthria.
本研究旨在：(a)比较有经验的言语语言病理学家（SLP）对构音障碍儿童和正常发育儿童的语音准确性和超鼻音感知判断的评分者间可靠性；(b)确定影响构音障碍儿童这些感知判断可靠性的语音因素。方法：通过网络平台，10 位 SLP 对 20 名 5 岁构音障碍儿童和 20 名 5 岁正常发育儿童的语音样本进行了超鼻音判断任务和语音准确性判断任务的评分。比较了构音障碍儿童和正常发育儿童的 SLP 评分者间可靠性。对于构音障碍儿童，四个声学语音测量指标、可懂度和语音准确性（正确停止百分比）被用来预测 SLP 感知判断的可靠性。结果：构音障碍儿童的 SLP 语音准确性判断和超鼻音评分的可靠性明显低于正常发育儿童。在构音障碍儿童中，感知判断的评分者间可靠性从强到弱不等。正确停止百分比是语音准确性判断和超鼻音评分的最强预测指标。结论：与正常发育儿童的评分相比，有经验的 SLP 对构音障碍儿童的感知语音准确性判断和超鼻音评分的可靠性降低。研究结果强调需要更可靠的方法来评估构音障碍儿童的语音准确性和超鼻音。
