Language for academic purposes : performance of grade-one English second language (ESL) learners on the diagnostic evaluation of language variation (DELV).

• Main Author: Kallenbach, Amy
• Format: Others
• Language: en
• Published: 2008
• Online Access: http://hdl.handle.net/10539/5674

This study endeavoured to document the oral English language abilities of grade-one, isiZulu-speaking English Second Language (ESL) learners, from three inner city schools using the Diagnostic Evaluation of Language Variation (DELV) (Seymour, Roeper & de Villiers, 2003). A cross-sectional quantitative design was used. This design consisted of both a comparative component, which examined performance and gender comparisons, and a correlational component that evaluated the relationship between the parent-teacher estimations of language proficiency, and the scores obtained on the DELV. In addition, there was a small element of qualitative research, which attempted to elucidate the children's general language environment. On the basis of the results obtained on the DELV, it was concluded that the groups of ESL children from the three schools would be treated as one larger group. The results from this larger single group showed that the ESL children in this study obtained low means and a wide range of scores for all three sub-tests of the DELV. This broad range of scores indicated that there is a large degree of individual variation amongst these ESL children. The placing of the scores obtained in each sub-test into their appropriate categories of weakness, low average, average, and strength, revealed that the ESL children in this study performed better in the pragmatics sub-test. This categorisation of scores was also utilised to detect impairment on the DELV. The results displayed that the majority of the ESL children's performances reflected a language impairment (LI) profile. In contrast, this strong pattern of LI was not clearly evident in the results from the item analysis in each sub-test. The results from the item analyses showed that the most difficult item for the ESL children in this study was the double wh-question (pragmatics sub-test), and that these ESL children experienced particular difficulty with the task of producing the correct articles and verb-contrasts. In addition, the results from the item analyses revealed these ESL children's learning potential. It is this learning potential that is the focus of dynamic assessment, which was recommended for the use with ESL children in further studies. The gender performances demonstrated that there were no differences between the male and female learners in this study across all three sub-tests. The generally weak relationships between the II DELV sub-tests and parent and teacher ratings of the ESL children's English proficiency, showed that although the DELV is a useful tool for the South African population, it should be used in conjunction with other language assessment tools (e.g. measures of language processing and language aptitude skills). The parent and teacher language questionnaires, which raised three issues: (1) The emphasis placed on English skills, as parents view it as providing an improved lifestyle for their children; (2) The effect that the ESL children's linguistic exposure in their L1 may have on their L2 development; and (3) The influence of the ESL children's linguistic experience in the classroom. Two main implications emerged from these results. First, is the need for further research on South African ESL children's English language acquisition, as well as for additional evaluations of the DELV's potential for this population. Second, is that the weak areas of English language skills displayed by the ESL children's results on the DELV must be addressed, by utilising the influence of L1 exposure in the home (the parents), classroom exposure (the teachers), and the expertise of the speech language pathologist (SLP).