Ptok M, Schwemmle C, Iven C, Jessen M, Nawka T
Klinik und Poliklinik für Phoniatrie und Pädaudiologie, Medizinische Hochschule Hannover.
HNO. 2006 Oct;54(10):793-802. doi: 10.1007/s00106-005-1310-6.
For routine clinical purposes, dysphonic voices are assessed perceptually using the GRBAS scale or analogues. For clinical application, the crucial question is the interrater reliability (IRR) of the auditory perceptual assessment of voice quality. Therefore, the IRR of the four point RBH (roughness, breathiness, hoarseness vs overall grade) scale was studied. Other parameters, e.g. validity and intrarater reliability were not considered.
A total of 78 patients read a standard text "Der Nordwind und die Sonne". These samples were evaluated by 19 speech and voice therapy students according to the degree of roughness, breathiness and hoarseness. Data were subjected to reliability analysis.
Our data indicate a high IRR with a Cronbach's alpha of 0.94. No single rating of the 19 raters could be omitted without decreasing the IRR.
The data indicate that the perceptual assessment of hoarseness for running speech is highly reliable. The application of the RBH scale is suitable for clinical purposes. It should be considered as an outcome measure.
在常规临床实践中，使用GRBAS量表或类似量表对嗓音障碍患者的声音进行感知评估。对于临床应用而言，关键问题在于声音质量听觉感知评估的评分者间信度（IRR）。因此，本研究探讨了四点RBH（粗糙、气息、嘶哑与整体等级）量表的评分者间信度。未考虑其他参数，如效度和评分者内信度。
共有78名患者朗读标准文本《北风与太阳》。19名言语和嗓音治疗专业学生根据粗糙、气息和嘶哑程度对这些样本进行评估。对数据进行信度分析。
我们的数据显示评分者间信度较高，Cronbach's α系数为0.94。若不减少评分者数量，19名评分者中的任何一人的评分都不能省略。
数据表明，对连续语音嘶哑程度的感知评估具有高度可靠性。RBH量表适用于临床目的，应将其视为一种结果测量指标。
