The case for bilingual language tests: a study of test adaptation and analysis
The justification for the use of language tests in education in multilingual and multicultural societies needs to include both the aims of bilingual education, and evidence that the international standards for tests that are available in two or more languages are being met. In multilingual and multi...
Saved in:
Published in: | Southern African linguistics and applied language studies Vol. 27; no. 3; pp. 301 - 317 |
---|---|
Main Author: | |
Format: | Journal Article |
Language: | English |
Published: |
Taylor & Francis Group
01-11-2009
|
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | The justification for the use of language tests in education in multilingual and multicultural societies needs to include both the aims of bilingual education, and evidence that the international standards for tests that are available in two or more languages are being met. In multilingual and multicultural societies, language tests, for example, can play an important role in the implementation and monitoring of additive bilingual education, or developmental maintenance bilingual programmes (MBE). MBE is contrasted with transitional bilingual education (TBE). One such programme of MBE for isiXhosa-speaking children in the rural areas of South Africa is currently being implemented. A test of cognitive academic language proficiency, the Woodcock Muñoz Language Survey, was selected to research the language outcomes of the project. This test is used widely in the USA for the evaluation of bilingual programmes, and is available in Spanish and English. After receiving permission for the test to be adapted into isiXhosa, a team of translators, educators, linguists and a psychometrist started the process of adapting the test into isiXhosa. In the language of the International Test Commission's Guidelines for Adaptating Educational and Psychological Tests (ITC, 2000), the challenge is to ensure the linguistic equivalence of the two versions of the tests, and to provide evidence of the statistical equivalence of the two versions of the test. The English version was also adapted for the South African context. This study will report on the processes that were followed during the workshops to translate and adapt the test, a Differential Item Functioning (DIF) analysis of the items of the two versions of the test, some of the steps that were taken to address problems in the initial adaptation of the test into isiXhosa, and the results of an evaluation of the construct equivalence of the two versions. |
---|---|
Bibliography: | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
ISSN: | 1607-3614 1727-9461 |
DOI: | 10.2989/SALALS.2009.27.3.8.942 |