Bibliography: leaves 168-181. / Interpreting is an act that has become so customary within the South African context that it has become virtually invisible. Most health care institutions rely on ad hoc interpreting, which involves haphazard interpreting arrangements in which anyone who speaks the patient's language, is called on to interpret. Untrained interpreters are consistently used in clinical practice due to a severe lack of trained interpreters. Despite this, to date little research has been conducted investigating the differences between trained and untrained interpreters. Furthermore, little research has been undertaken on the use of interpreters in the field of Speech-Language and Hearing Therapy. There is much need for interpreters within our profession, as clinicians usually either speak English or Afrikaans, with very little or no knowledge of indigenous South African languages. In this study, differences between trained and untrained interpreters were examined in the initial assessment interview, within the field of Speech-Language and Hearing Therapy.

Bibliography: leaves 62-69. / The spoken language productions of three language impaired learning disabled and non-learning disabled young adolescents were explored using a narrative discourse analysis procedure. The purpose was to compare the narrative production abilities of language impaired learning disabled and non-learning disabled young adolescents on a number of measures: story length, amount of information, coherence, evaluation and cohesion. In addition, it compared the performance of language impaired learning disabled and non-learning disabled young adolescents on three types of narrative tasks (story generation from a picture, personal narrative, and story retelling). Furthermore, this study investigated the stability of the narrative scores over three measurement times. The narrative discourse productions were analysed using more stringent reliability measures. Measures of reliability were obtained by calculating inter-rater and intra-rater reliability measures and testing for stability of scores across the three testing sessions. The results indicated that 46 of the 48 dependent measures in this study remained stable across the three testing times. One of the most important clinical findings in this study was that the language impaired learning disabled young adolescents are able to retell and generate narratives. Significant differences were however, found between the two groups. The language impaired learning disabled young adolescents produced shorter and less coherent and cohesive stories than the non-learning disabled young adolescents. Although they used fewer cohesive devices, they did not use significantly more incomplete and erroneous cohesive ties than the non-learning disabled young adolescents. The analysis of task effects indicated that story retelling is more clinically useful with language-impaired learning-disabled young adolescents for an assessment of narrative discourse ability than story generation. The implications for use of narrative discourse and types of story collection tasks for the assessment of language impaired learning disabled young adolescents are discussed.

The spoken language production of eight Afrikaans speaking Coloured persons with mild to moderate aphasia and eight Afrikaans speaking Coloured persons without aphasia were explored using a narrative discourse analysis procedure. The aim of the study was to investigate the discourse characteristics of the Cape Afrikaans dialect in persons with and without mild to moderate aphasia. Furthermore the study aimed to establish the cultural appropriateness as well as the clinical applicability of the newly developed Discourse Test Battery (Ulatowska et al., 1998) for the Afrikaans speaking Coloured population. In addition the adaptation features as well as the dialectal and discourse ethnic markers features were identified. The narrative Discourse Test Battery devised by Ulatowska et al., (1998) consisting of two composite pictures, a picture sequence story, a story retell and personal experience task was administered to all subjects. For the purpose of investigating the higher cognitive and linguistic processes, the formulation of main ideas, providing the lessons for the stories as well as proverb interpretation tasks were included and analysed. The narrative discourse productions were transcribed and analysed using more stringent reliability measures. Measures of reliability were obtained by calculating inter-rater and intra-rater reliability measures. The data was analysed in terms of the length of narratives, pr6positional units, quality analysis, analysis of evaluation analysis of dialectal features and ethnic discourse markers as well as adaptation features. The results indicated that for all of these methods of assessment the experimental group performed poorer than the control group. Statistically significant differences were noted on the measures of expressive phonology, lexical items, global structure, suspense, clarity and temporal sequence of some of the narrative tasks. One of the most important clinical as well as theoretical findings was the overlap between some of the dialectal features, general discourse features as well as adaptation features. Analysis of the task effects indicated that the picture sequence and the personal experience narratives showed the greatest differences between the groups and revealed the most information with regard to dialectal features, discourse ethnic markers and adaptation features. The spontaneous proverb and multiple choice proverb interpretation tasks revealed greater difficulties in abstracting and generalising information for the experimental group than for the control group. The assessment tool was found to be culturally appropriate for the Afrikaans speaking Coloured population of the Western Cape.

Early identification of hearing impairment in children is essential to avoid potentially disabling effects of hearing loss or deafness. This necessitates effective screening measures appropriate to the community in question. Current methods used in South Africa, especially for pre-school and school going children have resulted in poor coverage as they are designed for the more developed countries. There is thus a need to devise a screening method that is appropriate to our local conditions. In this study, a free-field live voice test was developed based on three levels: whisper, conversational and loud. This was evaluated against pure tone audiometry for sensitivity, specificity, cost and ease of application in two studies: hospital and school- based. A total of 394 children were tested; 189 in hospital-based study and 205 in school based study. 378 of the total were eligible for analysis. In the hospital-based study, the results of 177 children were analysed. The age range was 3 - 12 years with a mean of 5.8 years. The sensitivity (ability of the test to detect hearing impairment) was 80.0%; and the specificity (ability to identify children with normal hearing) was 95.0%. In the school-based study, done after modification and standardisation of the test set, the sensitivity and specificity were 83.3% and 97.8% respectively. Age range was 3 - 8 years with 79% being 4- 6 years. In both studies, the voice test was simpler to perform, easily understood and acceptable to the children and the testers; and considerably cheaper as the only equipment required was picture/toy set. The main limitation was non-standardisation of the test set. This was rectified in the school-based study. The drawbacks noted were the inability of the voice test to detect unilateral hearing loss/deafness and high frequency hearing loss. The voice test generally correlated well with pure tone audiometry and could be used as alternative for screening for hearing impairment in the community especially for pre-school and school going children. However, it is recommended to repeat the study in actual community settings using Community Health Care Workers as the testers. This would also determine the reliability of the voice test, as this cannot be reliably established at this stage.

The thesis is aimed at an analysis and automatic detection of logopaedic defects in speech utterance. Its objective is to facilitate and accelerate the work of logopaedists and to increase percentage of detected logopaedic defects in children of the youngest possible age followed by the most successful treatment. It presents methods of speech work, classification of the defects within individual stages of child development and appropriate words for identification of the speech defects and their subsequent remedy. After that there are analyses of methods of calculating coefficients which reflect human speech best. Also classifiers which are used to discern and determine whether it is a speech defect or not. Classifiers exploit coefficients for their work. Coefficients and classifiers are being tested and their best combination is being looked for in order to achieve the highest possible success rate of the automatic detection of the speech defects. All the programming and testing jobs has been conducted in the Matlab programme.