Abstract:
This study compares the different primary school language of instruction models found in South Africa for performance on the higher-order reading comprehension processes tested in the international, large-scale assessment, prePIRLS 2011. The language of instruction groupings include the African languages, Afrikaans, English L1 and English L2 as an immersion model. The marked differences in performance observed on the higher-order prePIRLS 2011 items across the language models are not singular reasons for performance in themselves, but reflect embedded contextual factors that influence performance. It was, therefore, examined how language of instruction creates or restricts access to ‘social tools’ and ‘cultural capital’ which mediate the development of higher-order reading comprehension at home and at school. It was found that socio-economic status (SES) at the school level contributes 86.06 (SE=20.48) and 98.54 (SE=17.28) score points for English (N=2 205) and Afrikaans (N=1 463) respectively as a school level indicator. For the Afrikaans LoLT grouping access to text explained 44 score points (SE=11.09). Neither SES nor access to text explained significant variance in performance for the African languages schools when examined within the language grouping (N= 12 076). A linear regression (N=6 342) showed that low SES learners whose language of instruction is English, despite it not being their mother tongue, benefit by 20.35 score points (equivalent to half a year) from being in the English L2 group, in comparison to the African languages L1 group as a measure of achievement on the higher-order subscale. The hypothesis that SES and access to text significantly affect performance in higher-order reading comprehension was further supported by the findings of a two-level regression, showing that access to text at school contributed 32.91 score points (B=32.91, SE=13.96, p=.03) to performance for English L2 low SES learners (N=480). It is argued that better provision of text at school can mediate the development of the cognitive and metacognitive reading strategies required for higher-order reading comprehension across all language of instruction models and socio-economic strata. Print material is often more accessible to learners in English. This strengthens a pragmatic, contextually-based argument for focused prioritisation of English L2 instruction concomitant to raising the quality of home language instruction.
