A research project submitted in partial fulfilment of the requirements for the degree of MA in Psychology in the Faculty of Humanities, University of the Witwatersrand, Johannesburg, 20 June 2018 / This study explores the relationship between multilingualism and working memory. Multilingual advantages in various executive functions have been established, but little is known about whether multilingual advantages extend to working memory capacity and functioning, or about the effect of speaking more than two languages. In a sample of 189 multilingual young adults in South Africa, this study used a multiple regression design in which numerous aspects of multilingualism - balance in proficiency across and within languages, the age of acquisition of additional languages, and speaking a third language - could be compared with one another while controlling for socio-economic status. Four aspects of working memory (verbal storage, verbal processing, visuospatial storage and visuospatial processing), measured using the Automated Working Memory Assessment (Alloway, 2007), acted as the dependent variables in respective regressions while independent variables measuring multilingualism, including the continuous measures of balance in reading, speaking and understanding proficiency across languages, were based on self-report information from the Language Experience and Proficiency Questionnaire (LEAPQ; Marian, Blumenfeld, & Kaushanskaya, 2007). Balance in proficiency emerged as a strong predictor of the verbal processing component of working memory, while no aspect of multilingualism significantly predicted visuospatial working memory. Combined with other results, this finding suggested that the effect of multilingualism on working memory may not follow the pattern observed in other tasks where multilinguals are advantaged in domaingeneral executive functions (like inhibitory control) but disadvantaged in linguistic tasks. Multilinguals’ experience in storing and processing linguistic information may lead to advantages (possibly through managing attention) that are specific to this kind of information. Keywords: bilingual advantage, executive function, multilingual advantage, trilingualism, working memory ! / GR2019

This study is a qualitative evaluation of how one state school interprets and makes a decision on the language medium to use as guided by the Language in Education Policy [LiEP] that advocates multilingualism in schools. The study asks how the LiEP ideal informs the language policy in the school, and establishes reasons parents give for choosing English as medium of instruction to be used in classrooms. Following Parlett and Hamilton‟s (1976) evaluation as illumination framework, this study outlines the language medium ideal expressed in LiEP and describes the actual Language Policy of the School in practice and how it accords with LiEP in guiding the medium of instruction. Data collection methods in this report included document analysis, classroom observations, interviews and questionnaires. The key findings from the data illuminate; parents of the learners in the school view the national language policy in a positive light that it is inclusive despite the challenges it presents to implementation. The findings reflected the decision makers‟ endeavour to strengthen their case that by choosing English as medium of instruction is not to contravene the policy, but a democratic right to benefit their children. Findings in the report suggest that the national language policy in South Africa is regarded a valuable document to guide the selection on the medium of instruction in schools, but raises issues that need to be addressed to make it play a more effective role in educational contexts.

Thesis (M.A. (Translation Studies and Linguistics)) --University of Limpopo, 2012 / The research reported in this mini-dissertation is an ethnographic case study which sought to investigate the development of biliteracy in one of the rural primary schools in the Limpopo province. Its focus is on how Grade 3 learners engage with texts and the strategies that teachers use to promote biliteracy (in English and Sepedi). Data collection methods included classroom observation, semi-structured teacher interviews and analysis of teaching and learning materials and the print environment. A brief analysis of the school’s language policy was also completed. The research revealed that the learners are hardly being taught to read and write whether in Sepedi, (their home language) or in English. While the school language policy states that English should be introduced in Grade 2, it is actually taught only in Grade 3. Additionally, as the learners do not understand English, the teachers frequently code-switch into Sepedi and therefore the learners hardly get any exposure to English. Many other negative aspects were uncovered. Out of the 28 lessons scheduled to be observed only 20 lessons actually took place. The learners are therefore not actually spending the allocated time on literacy development. The teaching is highly routinised with teachers, by and large, using an approach that emphasises repetition and rote-learning. The learners hardly ever get a chance to engage with texts independently. Even the textbooks available are not used but are stored away in the cupboards. Teachers painstakingly copy material from the textbooks on to the chalkboard and learners then copy this into their exercise books. The classroom environment is uninspiring, as there are hardly any learning materials on display. The interviews showed that the teachers had not been adequately trained to teach literacy and were in fact unaware of more effective ways of getting learners to engage with texts. They saw themselves as victims of frequent policy and curricular changes and blamed Government for poor training and lack of resources. The study in fact confirms findings of earlier research that the acquisition of literacy is simply not taking place in the poor, rural schools of South Africa and there is indeed a crisis in education in these schools.

Thesis (M.A. (Translation and Lingustics)) -- University of Limpopo, 2014 / The research reported in this mini-dissertation is a qualitative study, which sought to investigate the patterns of translanguaging in classrooms in the Foundation phase in a primary school in the Limpopo province. The aim of the study was to investigate the ways in which translanguaging is used by teachers and learners in the Foundation phase in a selected primary school in the Limpopo Province. The research focuses on how Grade 1 and Grade 3 learners and their teachers engage with texts and the strategies that teachers use to promote the use of two languages in classrooms to help learners to understand content and concepts in English and Sepedi. An innovative element of the research was the intervention teaching done by university lecturers to provide alternate practices for regular teachers in the school to discuss and engage with. The data collection instruments included classroom observations, audio and video recordings, interviews with the class teachers and a focus group discussion between the teachers. The data analysis involved identifying all instances of translanguaging that occurred in the lessons and to explore in what ways they facilitated learning. The results showed that hardly any translanguaging took place in the regular lessons and teachers seemed to be operating with a monolingual consciousness. Teachers also revealed in the focus group discussion that the Curriculum assessment and Policy Statement (CAPS) required them to keep the two languages apart and not to use both of them in lessons. In the intervention lessons, however, there were some examples of translanguaging, which seemed to facilitate interaction and greater participation from the learners. The mini-dissertation ends with some reflections on the findings, implications of the findings for future research and training, and recommendations to use the languages of school children as rich resources for teaching and learning.

The research described in this thesis is a case study of the communicative repertoires of teachers and learners of a grade seven class in a small rural primary school. The aim of the investigation was to find out if code switching is different in content-subject classes than in an English language class. The study accomplishes this by looking at code switching in different learning areas. Different research techniques such as observations, video. recording, taking of field notes and interviews are used as the means of data collection. The study does not attempt a full linguistic description of the switches made by the research subjects. For example, it does not deal with linguistic constraints on code switching, but instead it focuses on the functional aspects of code switching. The study reveals that there are various functions of code switching in the classroom. The following are the major findings noted in this study: • Code switching is a communicative resource, which enables the teachers and learners to accomplish a considerable number and range of social and educational objectives. • Code switching is used to negotiate and renegotiate meaning. • How teachers code switch in class depends on how difficult the lesson is for the learners. • The study reveals no major differences between code switching In content subjects than in English as a subject. • Although tbe subjects engage in code switching, they seem to feel guilty about their behaviour.

The number of African learners who participate and succeed in physical science is recognized to be ‘disturbingly low’. One of the factors attributed to the low levels of enrolment and performance in physical science is language. In this study, teaching and learning through the language of science is examined in the context of the bilingual classroom. A model of analysis is constructed that (1) extends the notion of the language of science to include the mathematical and visual ‘languages’ of science, (2) takes recognition of the manner in which language, content, and values and beliefs construct the science learner, and (3) moves beyond the characterization of teaching and learning according to the dichotomy of the ‘traditional’/ the ‘progressive’. The model of analysis draws upon the central concepts of a sociocultural model of pedagogy, namely the ‘developmental model’. In addition, the model of analysis makes use of Systemic Functional Linguistics to examine teaching and learning at the micro level of classroom interaction. This study reveals the complex nature in which the language, content, and values and beliefs change as a lesson unfolds: teaching and learning through the language of science has been shown in these classrooms to be marked by features of both a ‘traditional’ and a ‘progressive’ model – each of which appears to serve different functions in the overall construction of the science learner. In addition, this study begins to uncover how a ‘successful’ teacher equips his/ her learners in the context of the bilingual physical science classroom: teaching and learning through the language of science has been shown in these classrooms to incorporate complex and varied strategies that depend upon choices made by both the teacher and learners. These findings substantiate the need to understand the challenges teachers and learners face in the bilingual physical science classroom in ways that acknowledge the complexity of the teaching and learning process.

Research on code switching (CS), the use of more than one language in a single piece of discourse, has focused on various aspects of the phenomenon. For example, research has concentrated on switching that occurs between turns of interlocutors in a piece of discourse, sometimes between sentences within a single tum and sometimes within a sentence. Researchers have approached this discourse behaviour from various perspectives. For instance, some investigate the social functions of the switches, others explore the linguistic constraints on the switches. Furthermore, most of this research has examined CS in non-educational contexts. Research on classroom CS, the focus of this study, took hold in the mid 1970s. Researchers began to investigate the communicative functions of CS and the frequency with which teachers and learners used certain languages to perform different functions. Recent studies on classroom CS focus on the sequential flow of classroom discourse and "the way in which codeswitching contributes to the interactional work that teachers and learners do in bilingual classrooms" (Martin-Jones 1995:91). The approach used in these studies is the conversational analytic approach grounded in ethnographic observation. In South Africa little research has been done on classroom CS, though it makes an important contribution to the interactional work of teachers and learners in classrooms. This study explores the use of English and Xhosa in the classroom as a learning and communicative resource. Its focus is on the communicative functions of the switching behaviour of a teacher as she interacts with her pupils. In other words, this study looks at how the research subject uses English and Xhosa to get things done in the classroom. As the classroom situation observed is dominated by the teacher, this study concentrates mainly on her communicative repertoire. It does not attempt a full linguistic description of the switches made by the research subject, for example, it does not deal with linguistic constraints on CS. One of the major findings noted in this study is that the research subject does not use CS so as to avoid using English. She uses CS as a learning resource. It has also been noted that CS is used by the research subject as a contextualisation cue, for example, we noted the way it co-occurs with other contextualisation cues like nonverbal communication cues. This is in line with Martin-Jones' (1995) viewpoint that CS is not used by bilingual teachers simply to express solidarity with the learners but to negotiate and renegotiate meaning.

Thesis submitted in fulfilment of the requirements for the degree Master of Education in the Faculty of Education at the Cape Peninsula University of Technology 2014 / This study investigates the effects of code switching in the learning of mathematics word problems in Grade 10. The research used Cummins’ language acquisition theory to inform the study. The study employed ethnographic qualitative research design whereby classroom observations and semi-structured interviews were used as data collection techniques. The use of multiple data collection techniques was to ensure validity and credibility of the study. The sample consisted of sixty learners and two mathematics teachers. The sample was drawn from a population of one thousand two hundred and thirty five learners and forty nine teachers.The study has shown that even though code switching could be beneficial in the learning and teaching of mathematics, it was difficult for learners and teachers to use it in a way that enhances the learning of mathematics word problems because of the barriers in the use of mathematical language. It is recommended that teachers should exercise care when using code switching, especially with the topics that involve word problems; as such topics are more aligned to certain mathematical language that could not be translated to IsiXhosa.

A research report in part fulfilment of the requirements of an MSc in Science Education at the University of the Witwatersrand / This study reports the findings of a research project in a DET (Department of Education and Training)* school in Mamelodi, South Africa, in 1995. An investigrtion was carried out to ascertain whether students benefit from having access to science definitions in their vernacular, and to determine their preference in terms of the medium of instruction in the science classroom. A Science Concept Guide, written at RADMASTE** Centre, was used as the main component of the intervention in a pretest - post-test study. This was followed by student questionnaires and interviews. The Concept Guide consisted of definitions and explanations of Physics and Chemistry concepts arranged according to themes. These explanations were translated into North Sotho. Two groups of Std 10 students participated in the evaluation of the Concept Guide. One group used an English version of the Concept Guide and the other an English and North Sotho version. The differences between the pre-test and post-test scores of each group were compared. The results indicated that a significant improvement in scores had occurred in the English and North Sotho Group. Interview results showed that a number of students were in favour of having learning material of this sort available in their vernacular and claimed that they had found it helpful. In general, both groups expressed favourable comments about the Guide. Opinions about medium of instruction varied. Overall, findings indicate that further efforts in the development of vernacular material for the learning of science would prove beneficial. Although the area of policy decision is outside the scope of this study, it hopes to contribute in some way to the knowledge about the preferred medium of learning in South Africa. / Andrew Chakane 2018

This is a study of secondary mathematics teachers' knowledge of the dynamics of learning and teaching mathematics in multilingual classrooms in South Africa. It probes teachers' articulated and tacit knowledge through a qualitative methodology that includes In-depth interviews, classroom observations, and reflective workshops. The sample is purposive and theoretical, comprising SIX teachers drawn from three different multilingual school contexts. Categories of description and analytic narrative vignettes enable a qualitative, layered analysis of what the teachers said and how they acted.