Pattern Of Pronunciation And Writing In Developing Somali Braille, Thika, Kenya

ABSTRACT

This study was intended to establish the patterns of pronunciation and writing for Somali Braille because a Standard Somali Braille does not exist. It examined, identified and hypothesized the similarity of speech sounds, grapheme symbols and Braille codes, based on experiment and a 3x2 factorial design. The independent variables were Braille codes and Somali print, while the dependent variable was Somali Braille. The population were students, who were blind (112) and low vision (147), totalling to 259 and a sample size of 90 participants was used. The instrument was a questionnaire with questions consisting of the following speech sounds- English 26, Kiswahili 35, Arabic 30 and Somali 31. Somali sounds were produced by the instructor during the experiment. Each question had two options: A) Pronounced and written same way and B) Pronounced and written differently. For primary analysis, Yule’s correlation coefficient was used to interpret data involving HO1, HO2 and HO3 about the significant relationship of the speech sounds, grapheme symbols and Braille codes in English, Kiswahili and Arabic and Somali print. HO4 was tested using the chi-square test. Validity was tested by chi-square and reliability was tested by instruments used to measure the consistency of results. The result obtained was a total score of 1815 (63%) for similarity and 1065 (37%) for difference against a total score of 2880 (100%). That means 58 (64.4%) and 32 (35.6%) of participants agreed and disagreed respectively. Using Yule’s coefficient of association value between Braille codes and Somali print was found to be +1. That means the class frequency of Braille code and Somali print was greater than the expectation of Somali Braille. In testing the null hypothesis, the table value of χ² was 5.991 and calculated value of χ² was 1.777 for 2 degrees of freedom at 5 per cent level of significance. That means the calculated table is much lower than this table value and hence the result of the experiment supports the null hypothesis and proves that the Braille code and Somali print are independent variables. The research findings indicate there is a positive and significant similarity among the speech sounds of English, Kiswahili, Arabic and Somali languages. This implies that the existing Braille codes for English, Kiswahili and Arabic can be used to develop Somali Braille. The recommendations include the development of Somali Braille contractions; establishment of Somali Braille centres; the governments, international community and well wishers to make efforts to come up with a strategy of developing Somali Braille letters and modifying the existing Braille code that is acceptable to Braille users.