A thesis submitted to the Faculty of Humanities,
Department of Psychology at the University of the Witwatersrand,
in fulfilment of the requirements for the degree of Doctor of Philosophy
October 2016. / Literacy in multilingual contexts includes social and cognitive dimensions
(GoPaul-McNicol & Armour-Thomas, 1997). Becoming literate carries with it the ability to develop
and access higher-order thinking skills that are the building blocks for cognitive academic language
proficiency, as well as the means that define educational opportunities (Bialystok, 2007). South Africa
has 11 official languages and a multilingual education policy but South African schools are able to
determine their language of instruction policy of monolingualism or multilingualism (Heugh, 2010).
This raises the question of whether monolingualism or bilingualism influences children’s successful
acquisition of reading. It is important to investigate the effect this has on reading processes and skills
of monolingual and bilingual children because this issue has received limited research attention while
it contributes to our greater understanding of how children’s cognitive capacities for literacy
attainment are either constrained or promoted through broader social factors operating in a child’s
literacy-learning environment (Bialystok, 2007; Vygotsky, 1978). Cognitive processing and reading
skills were assessed in monolingual and bilingual children at a public school in an urban area of
Johannesburg. An English-speaking monolingual group with English as the language of instruction (N
= 100) was compared with a Zulu-English bilingual group with Zulu as first language (L1) speaking
proficiency and English as second language (L2) literacy experience (N = 100) on measures of
reading, phonological awareness, vocabulary skills, and working memory. Performance in cognitive
processing and reading skills of these two groups was compared to an Afrikaans-English bilingual
group (N = 100) with dual medium instruction. Tests of language proficiency confirmed that the
Afrikaans-English bilinguals were balanced bilinguals and that the Zulu-English bilinguals were
partial bilinguals.
Aim and method: The purpose of this study was to expand knowledge in the field of second
language reading acquisition and language of instruction by examining the impact of language related
factors on the cognitive development and literacy competence of monolingual and bilingual children
in the South African context. The central tenet of the bio-ecological approach to language, cognitive
and reading assessment is that language acquisition is inseparable from the context in which it is
learned (Armour-Thomas & Go-Paul-McNicol, 1997). Drawing from this approach, the present
research project investigated the effects of the level of orthographic transparency on reading
development in the transparent L1 and opaque L2 of biliterate Afrikaans-English bilinguals learning
to read in a dual medium school setting. The effects of oral vs. written language proficiency in the L1
on the acquisition of L2 English reading was also investigated by examining whether reading
processes and skills transferred from one language to another and the direction or nature of this
transfer in partial and balanced bilinguals. Finally, whether a balanced bilingualism and biliteracy
Cognitive processing skills and literacy development in monolingual and bilingual children in South Africa
vi
experience had beneficial effects on cognitive tasks demanding high levels of working memory
capacity, was investigated.
Results: Reading in Afrikaans – the more transparent orthography – reached a higher
competency level than reading in the less transparent English. Dual medium learners and L1 English
monolingual learners acquired reading skills in their home language(s) at a higher level than L2
English with L1 Zulu speaking proficiency learners did. Dual medium learners outperformed both
monolingual learners and L2 English with L1 Zulu speaking proficiency learners on tests of
phonological awareness, working memory, and reading comprehension. They also reached similar
competency levels in tests of vocabulary knowledge than monolingual English (L1) learners. These
differences translated into different relationships and strengths for reading attainment in monolingual
and bilingual children. These findings provide support for a language-based and context-dependent
bio-ecological model of reading attainment for South African children.
Conclusions: Bilingual children who are exposed to dual medium reading instruction
programmes that value bilingualism philosophically and support it pedagogically create optimal
conditions for high levels of cognitive development and academic achievement, both in the first and
in the L2. Absence of mother tongue instruction and English-only instruction result in a reading
achievement gap between emergent Zulu-English bilinguals and English monolinguals. This effect is
not observed in the biliterate Afrikaans-English bilinguals; instead, these children performed better
than the English monolinguals on many English tasks and working tasks requiring high levels of
executive control and analysis of linguistic knowledge, despite English being their L2 while learning
to concurrently read in Afrikaans and English. Arguments for and (misguided) arguments against dual
medium education are examined to identify the consequences of translating this model of education
into effective schooling practices, given the socio-political contexts in which educational reforms take
place at local schools and in communities (Heugh, 2002). More broadly, good early childhood
education includes a rich language learning environment with skilled, responsive teachers who
facilitate children’s literacy learning by providing intentional exposure to and support for vocabulary
and concept development. Classroom settings that provide extensive opportunities to build children’s
reading competences are beneficial for young dual language learners no less than for children
acquiring literacy skills in a one-language environment (Cummins, 2000; Heugh, 2002). / GR2017
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:wits/oai:wiredspace.wits.ac.za:10539/22338 |
| Date | January 2016 |
| Creators | De Sousa, Diana Soares |
| Source Sets | South African National ETD Portal |
| Language | English |
| Detected Language | English |
| Type | Thesis |
| Format | Online resource (xiv, 405 leaves), application/pdf, application/pdf |
Page generated in 0.0031 seconds