Speech audiometry is a diagnostic hearing test used to evaluate a person’s ability to perceive, recognize and understand speech. At the moment, in Slovenia there is no suitable test available for speech audiometry in young children. In the thesis I therefore attempt to construct a monosyllabic speech audiometry test to be used for supra-threshold speech perception assessment in children from five to six years of age.
The test was developed in two phases. In the first phase of test construction, a new set of words suitable for children of selected age group was compiled. Part of the ability to perceive speech is filling in the missing speech information based on language knowledge; therefore, it is important that the test uses words that are familiar to the subjects. Because my goal was to construct a closed-set test, illustrations of selected words were also created in this phase. Familiarity with words and adequacy of their pictorial representations were evaluated in a pilot study on 20 normal hearing children. In the second phase, four phonetically balanced lists of approximately equal difficulty were created from the words that were found to be suitable in the first phase of test construction. The test was then evaluated on a group of 10 normal-hearing and a group of 6 hearing-impaired children.
As expected, the results of the pilot test evaluation showed a significant difference in average test scores between the two groups of subjects, indicating adequate construct validity of the test. Internal reliability was found to be good in the group of normal-hearing children, but not in the group of hearing-impaired children. The new monosyllabic test for children should therefore be further evaluated on a larger sample of participants and improvements of its metric characteristics are needed.
|
