This study was performed to standardize the Korean bisyllabic word lists for juveniles and preschoolers, which are used for the speech recognition threshold (SRT) test. A set of 50 bisyllabic words for juveniles and 24 words for preschoolers were first selected from previous literatures based on the familiarity, phonetic dissimilarity and normal sampling of Korean speech sounds. Then psychometric function tests for each word were performed for 10 juveniles and 10 preschoolers with normal hearing to choose a more homogeneous group of bisyllabic words with respect to audibility. All test words were randomly presented to each subject at each intensity level from -5 dBHL to 25 dBHL in 5 dB steps for juveniles and -10 dB HL to 30 dB HL in 10 dB steps for preschoolers.
For each group, the percentage of correct responses was plotted as a function of intensity level. The slopes and levels at 50% intelligibility of the psychometric function curves for all test words were obtained using the logistic regression equation. The mean slope and threshold were 10.96%/dB (SD 1.63) and 7.89 dBHL (SD 2.48) for juveniles and 6.65%/dB (SD 1.83) and 13.28 dBHL (SD 4.62) for preschoolers. The standard deviation of the intensity in each spectrum of speech sounds was within 10 dB. Based on these results, we finally selected 24 bisyllabic words for juveniles. For preschoolers, we selected 12 words and made a board of
pictures composed of 3 rows and 4 columns each of which represents a bisyllabic word. To sum up, these bisyllabic word lists developed in this study are homogeneous with respect to audibility. Before being widely used to hearing-impaired children, however, it may be necessary to verify and revise these lists.
KEY WORDS:Bisyllabic word lists·Familiarity·Homogeneity with respect to audibility·Psychometric function curve· Speech recognition thresholds