Investigating the Inter-rater Agreement of Farsi-speaking Raters on the Grade, Roughness, Breathiness, Asthenia, and Strain Auditory Perceptual Scale

Abstract

Background: Auditory perceptual assessment is one of the important evaluations for voice assessment. Among the available auditory-perceptual assessments, the grade, roughness, breathiness, asthenia, strain (GRBAS) auditory perceptual scale has attracted the attention of many researchers and therapists in Iran and other parts of the world. The GRBAS scale is a perceptual voice assessment scale that subjectively assesses voice quality; however, the audio samples in the original GRBAS scale that are used as training tools for raters are Japanese. There are various segmental and suprasegmental differences between the Japanese and Persian languages since these factors can affect the accuracy of the perceptual evaluation. Objectives: This research aims to investigate the inter-rater agreement among the Persian raters who only had access to the Japanese samples in the main profile. Methods: In this study, 8 speech and language pathologists were selected as raters with more than 5 years of clinical experience in evaluating and treating voice disorders. Several 137 audio samples, containing the prolongation of vowel /a/ and reading the standard text “Grandfather Passage” were provided to the participating raters. The raters were asked to score the audio samples based on the GRBAS auditory perceptual scale. The results were statistically analyzed via the Cohen kappa coefficient. Results: The highest agreement in the R parameter was related to rater number 5 (kappa=0.585) while the lowest value was related to parameter S, related to rater number 1 (kappa=-0.018). Conclusion: According to the results, the clinical experience of raters using the Japanese samples cannot lead to an increase in the agreement and ability of Persian-speaking raters in evaluating Persian samples.