PURPOSE: The purpose of this study was to investigate whether language proficiency would have an effect on how listeners rate the severity of hypernasality.
METHODS: The study included two groups of listeners: 12 Cantonese native-speakers in Hong Kong and 12 English native-speakers in Edinburgh. All of them were speech and language therapy students. Both listener groups underwent a training on perceptual judgement of hypernasality before rating 252 non-nasal Cantonese sentences, using direct magnitude estimation. The non-nasal sentences were produced by 22 Cantonese speakers with hypernasality (11 male, 11 female).
RESULTS: The Cantonese listeners showed a statistically significant higher mean intrajudge reliability (r = 0.55) than the English listeners (r = 0.39) for rating male speakers (p < 0.05). There was no significant difference between the two listener groups (r = 0.63 for Cantonese listeners; r = 0.56 for English listeners) for rating females (p > 0.05). The interjudge reliability was high for rating females for both Cantonese listeners (0.91) and English listeners (0.88), using Cronbach’s alpha. For judging male speakers, Cantonese listeners showed high interjudge reliability (0.87), while the English listeners showed moderate reliability among judges (0.77). Furthermore, the perceptual ratings of both male and female speakers assigned by the Cantonese listeners were significantly higher than those assigned by the English listeners (p < 0.05).
CONCLUSION: The results suggested that listeners who are naïve to a language tend to be more conservative than the native speakers when they perceptually judge hypernasality. This could possibly affect the results of treatment outcome measure and the accuracy of diagnosis.