Some strategies used by isiZulu-speaking learners when answering TIMSS 2003 science questions.
The purpose of this study was to describe the performance of the South African Grade 8 learners in Trends in International Mathematics and Science Study (TIMSS) 2003 science test, to explore the translatability of TIMSS 2003 science items into isiZulu without significant loss of meaning, and to explore the strategies used by isiZulu-speaking learners when answering questions in the TIMSS 2003 test. Thirty six isiZulu-speaking learners were tested using written test questions taken from the science test in the TIMSS 2003. The degree to which a sample of 36 learners represented their understanding of the questions in a written test compared to the level of understanding that could be elicited by an interview is presented in this study. The findings of this study are presented, interpreted and discussed using Pollitt & Ahmed's (2001) model of question answering process as well as other relevant literature. The key findings of this study are as follows : • the South African Grade 8 learners performed very poorly on TIMSS 2003 science test, • close translation of TIMSS 2003 science items into isiZulu is possible if conducted with care by expert teachers, • the language of the test had some effect on isiZulu-speaking learners' performance on TIMSS 2003 science test, • the strategies used by isiZulu-speaking learners when answering science questions included: • translating the question into isiZulu before trying to answer it, • choosing an answer containing a word/term common in the question stem and in the options, • choosing the answer containing a familiar/unfamiliar word in the options, •guessing , •looking at patterns of previous choices, •'picture memory', and •'general knowledge'. When Pollitt & Ahmed's (2001) model of question answering is applied to isiZuluspeaking learners, two 'new' phases are introduced. The findings of this study suggest that language factors are embedded within other factors, importantly, the appropriate level of cognitive proficiency to enable correct answering of science questions. The findings of this study further suggest the need for development of cognitive/academic language proficiency (CALP) in both English and isiZulu languages, or in one of them.