This thesis explores the efficacy of think-aloud protocols (TOLs), and the extent to which the TOL technique is able to reveal the nature of ESL reading interaction with expository prose. The investigation constituted a critical evaluation of current ESL TOL research, which was essentially a theoretical examination of emerging problems derived from an in-depth assessment of current ESL TOL studies. The theoretical examination was supplemented by the practical implementation of the technique in a case study research, utilising three verbal protocols obtained from ESL students at the University of Fort Hare. The close observation afforded by the evaluative case study research paradigm provided the writer - as participant observer - with a further means of judging the merit of TOLs, which corroborated findings from the theoretical evaluation, and enabled a consideration of unanticipated issues which emerged from the practical implementation of the technique. The conclusion was that TOLs seem to have a unique ability to identify aspects of the nature of ESL reading gained from an on-line assessment of reader interaction, provided that TOL research is conducted within certain methodological and analytic research constraints. The writer has proffered suggestions for future ESL TOL research, and feels that the combination of TOL research findings with other measures of reading comprehension could elucidate aspects of ESL comprehension, making a valuable contribution to ESL reading theory and practice.

This study is a description of the language experiences of English Second Language students in their first year at Rhodes University. It took place in the context of the changes that are currently occurring in higher education in South Africa in terms of student populations. More and more students are entering tertiary education institutions, including HWESUs, such as Rhodes University, who are considered non-traditional. These students typically have English as their second, or additional, language, and have not been adequately prepared for university study by their secondary education. This study describes the experiences of three such students in their first year at Rhodes University. Entry into a university is seen not just as acquiring knowledge, but as entering, or attempting to enter, a new culture. It is recognised that all students enter universities with other cultures or literacies already in place. In the case of non-traditional students tbese other literacies are usually at some distance from those of the university. The work of James Gee (1990) is particularly useful in understanding this process of adjusting to the demands of university study and the effect that previous experiences have on this process. This study is an attempt to discover and describe the literacies that these three students brought with them to university and the effect these literacies had on their attempts to enter academic discourse communities of the university. An ethnographic research method was adopted in order to do this. The study is also an attempt to evaluate, from the perspective of the three students, the appropriacy of the various changes that Rhodes University has made since the numbers of non-traditional students has started to increase.

The purpose of this study was to analyse formative assessment challenges facing English Language (L2) Secondary school teachers in the Makoni District of Zimbabwe. Data were collected from 25 English Language secondary school teachers pooled from 5 secondary schools in the Makoni District. The study utilised the pragmatic paradigm which allowed the use of the mixed methods approach. The study used the survey research design. Quantitative data were collected through questionnaires while qualitative data were collected through follow-up interviews, documents and non-participant observations. Cross-tabulations were used to present data which were then reported mainly in percentages. While most of the qualitative data were used to buttress findings established through the questionnaires, the other data were categorised into themes and analysed accordingly. Major challenges that were revealed by this study were that teachers used the teacher – centred approach, feedback given to pupils was not detailed, there was less time to assess appropriately and that teachers were not motivated to assess effectively. Remedial activities were not being carried out and that teachers’ training in assessment was not thorough. Shortage of teaching and learning resources and high teacher – pupil ratios were some of the challenges that teachers faced. There was lack of coordination of agencies involved in assessment and that assessment was examinations – oriented. Assessment policy formulation did not involve teachers and that most pupils were not motivated to learn. On the basis of these findings, the study recommended the provision of adequate teaching and learning resources, provision of appropriate pre-service and in – service training programs as well as involving the teachers in the formulation of assessment policies.

Prompted by increasing demand in South Africa for the development of a focused but flexible English Second Language (ESL) curriculum at university level, this thesis contends that substantial theoretical under-pinning is needed for decisions on ESL course materials. Once the theoretical constructs are determined, a model based on a systematic approach to course design is proposed. It maximizes the individualization of experiential learning, despite the large numbers of students who take these courses, through a multi-form course structure offering four streams of study at three levels of difficulty. Entry is possible at the start of the year and at mid-year. The empirical research which forms the basis of the study is an analysis of the 1985 student group at the University of South Africa (UNISA). Several methods are used, including post-course questionnaires, diagnostic assignments and a detailed language and stylistic error count linked with a clause analysis of a sample of assignments and examination scripts. The model curriculum meets the contextually basic science requirements of a university course, within the parameters of response needed in regard to the ESL student profile determined by the needs and role analysis completed in Chapter 2. Model aims and terminal learning objectives are presented in Chapter 3 as the foundation on which the rest of the thesis is constructed, and include comprehension, applied composition, oral and aural skills, use of reference works, methods of thinking, and occupationally relevant specialist language. In Chapters 4 and 5, in-depth analyses of appropriate course content and methods emphasize the use of Afrocentric English literature in contemporary settings with appropriate readability levels, language in use in specified contexts, development of vocabulary, remedying incorrect usage, comprehension skills, composition skills, development of cognitive processes, oral and listening skills, and the purpose and place of grammar. The final chapters outline approaches to criterion-referenced assessment and evaluation, and suggest appropriate set works and criteria for their selection. The course materials aim at improving English communicative performance. The underlying principles used in developing this course design and its associated materials can be valuably extrapolated and applied at universities and other tertiary institutions. / English Studies / D. Litt et Phil. (English)

This thesis investigates the construct of productive vocabulary knowledge with a productive vocabulary task, Lex30. The task is designed to elicit up to four vocabulary items in response to each of 30 cues. In this way, Lex30 generates a corpus for each subject up to 120 words, which is then categorised according to frequency bands. The test is scored according to the number, or proportion, of infrequent words elicited, with infrequent defined as all items excluding the most frequently occurring 1000 English words. The higher the Lex30 score, then, the more infrequent words that subject has produced in response to the cues. Each corpus generated by Lex30, therefore, offers information about subjects' relative knowledge of infrequent items, although this might only be threshold knowledge. A feature of Lex30 is that it appears to measure productive vocabulary knowledge discretely: it does not activate multiple aspects of language knowledge, and is not context engaging. This feature suggests that we can measure one of the many aspects that are commonly considered to constitute language knowledge, productive vocabulary knowledge, without interference from other aspects of language knowledge. Additionally, Lex30 offers the potential to hypothesize about subjects' relative L2 proficiency in terms of the proportion of infrequent items they provide. To investigate the construct of productive vocabulary with Lex30, this thesis examines, in a principled way, exactly what aspect of language competence it measures, and makes comparisons with other cognate tests. The test has been used in a number of contexts since its introduction; this thesis offers a thorough investigation of its reliability, different versions of the scoring system, the influence cue frequency and of specific cue items, and the mode of task delivery and response. The thesis concludes that Lex30 provides us with a helpful means to understand the construct of productive vocabulary knowledge.

This study unravels the connector patterns and mental activities that the Chinese learners of English as a Second Language (ESL) and native-English speakers display/perform upon the adoption of adversative, causal and temporal connectors while writing English argumentations, descriptions and narrations. It focuses on the ways in which the Chinese ESL learners’ thinking and connector production recursively interact in meaning-building processes. To elicit information of the writers’ mental processes, three English writing tasks with concurrent verbal reports were used as main elicitation tools. Chinese university students from Low, Mid and High proficiency levels and native-American-English users (Native) were asked to speak out their thoughts while writing English argumentations, descriptions and narrations. Data were also elicited from the retrospective verbal reports and following-up interviews which were conducted right after each writing and concurrent verbal report. Comparison of connector patterns in English texts produced by the Chinese and native-English writers suggests that the Chinese Mid and Low groups use significantly more connectors than the Native group while connector frequencies between the Chinese High and the Native group reveal no significant difference. It has been argued that the High-level students, being more linguistically skilled, can use connectors as well as other cohesive devices to form textual connections as their native-English-speaking counterparts. However, the Mid and Low-level writers, lacking abilities of manipulating other cohesive devices, tend to rely mainly on connectors. Moreover, comparison between the three Chinese groups indicates that the Mid uses significantly more connectors in English texts than the High and Low groups. The Mid group, on the one hand, is not linguistically mature enough to use various cohesive devices as skillfully as the High group, but, on the other hand, has more knowledge on the connector usage than the Low. When connectors were analyzed by category, it is found that the Chinese students use fewer adversative connectors in argumentations and narrations than the native-English writers. This pattern is found to have arisen from the Chinese students’ rigid textual development. ESL learners, due to their poor second language (L2) proficiency, cannot develop sophisticated argumentative and narrative structures. Instead of resorting to counter-arguments and reverse-order narration, they tend to simply list ideas. This could result in lower occurrences of adversative connectors in their argumentations and narrations. The data that are extracted from concurrent verbal reports reveal that the Chinese students use their first language (L1) for five mental activities: (1) initiation, (2) revision, (3) comment-making, (4) decision-making, and (5) monologue-conducting while using L2 for only two: initiation and revision. Examinations on the cognitive demand of the mental activities show that ESL learners rely more on their L1 to perform the cognitively more demanding activities such as comment-making, decision-making and monologue-conducting. Furthermore, the verbal data show that the L2 proficiency levels affect ESL writers’ mental activities in two ways. First, proficient ESL learners report larger linguistic units than the less proficient ones. Second, proficient ESL learners report higher proportions of cognitively more demanding activities than the less proficient ones. These two patterns are argued to result from the different working memory capacities of the proficient and less proficient ESL writers as well as the different ways of processing information in their working memory. Finally, a model of L2 writing process is developed to delineate how L2 proficiency, working memory and cognitive process interact and contribute to L2 writing process. Theoretically, this L2 model refines the L1 model and enriches the current understanding of the L2 writing process in three ways. First, mental activities are classified according to how much cognitive workload they demand. Such categorization is necessary because it explains ESL learners’ language choice while doing verbal reports and also explains the difficulties the learners encounter in L2 writing. Second, the L2 model separates the composing process as pre- and post-stage, which allows us to see how frequencies and types of mental activities differ in the two phases. Third, by including language choice, the L2 model addresses a wider interpretation of the mental activities in writing process. Pedagogically, the comprehensive and explanatory accounts of the connector patterns and mental activities address the problems that ESL learners have when using connectors in L2 writing. This will definitely facilitate the teaching of L2 writing.

M.Ed. (Language Teaching) / Please refer to full text to view abstract

This study is an investigation of instances of conversational failure in interaction as evidenced by speakers of Black South African English (BSAE) , with a particular focus on the role of prosody in conversational (a)synchrony. The data analysed consist of six conversations, one SAE-SAE (South African English) encounter, four BSAE- SAE encounters and one BSAE- BSAE encounter. After a theoretical framework is set up, the analysis is conducted by means of two triangulation research processes based on Ethnomethodology. The analysis consists of an investigation into selected extracts which participants and informants alike perceived as 'stressful'. An attempt is made to isolate the sources of each instance of pragmatic failure. Prosodic features are found to be important in establishing and maintaining theme and conversational synchrony. But other factors are also involved. The analysis reveals two major influences of asynchrony: deviance in the use of (in order of importance) prosodic, lexical and syntactic cues to discourse functions; and a mismatch in the application of socio-cultural principles guiding conversational behaviour. The study leads into a brief outline of aims, objectives and conversational competence at a tertiary level and concludes with suggestions for further research.

South African education has been going through a process of transformation: from a traditional, transmission system to one of a more progressive nature. In the new curriculum, Curriculum 2005 anchored on Outcomes-based Education (OBE), focus is laid on critical reflection, innovation and creativity by classroom participants. Learning situations are expected to promote flexibility, co-operation and relevance of teaching/learning materials. This research looks at these aspects in the English Second Language (ESL) situation where print media articles are used as resource. Other than being a rich resource for language, print media texts have been found to be appropriate for this exploration because of their potential to offer opportunities for critical reflection and interaction with authentic issues. Unlike uncritical reliance on textbooks, the proper use of media articles can give the teachers and learners the space to select relevant and exciting materials for their situations. In this study, the research was done through action research where I (the researcher) actually taught the lessons in collaboration with the English teacher. The teaching method was influenced by the Freirean style of Iiberatory teaching where critical reflection, socio-political relevance and co-operation are crucial elements. The action research was done in two cycles. Experiences in the first cycle pointed to, among other aspects, the fact that the learners could not work in a co-operative way as they were not used to it. Addressing this problem consequently became the basis of the second cycle where the skills of co-operative learning were taught before the learners engaged in further lessons. The conclusions drawn from this research point to the importance of the teacher as an agent in the learning process if the demands of progressive and liberatory teaching are to be met. It has thus been concluded that elements such as critical thinking, creativity, flexibility and selection/development of relevant materials presuppose the existence of a well-trained teacher who is skilled and able to make these possible. In the light of my findings, my major recommendation is that teacher development needs to be strengthened so that more emphasis is placed on enhancing the teacher's ability to critique texts and improve his or her own practice, acquiring the skills to teach in a critical way and achieving the learning competencies.

This thesis analyses the way in which placement testing is being interpreted in Adult Basic Education and Training (ABET). The thesis examines whether the placement tests used in the case study were valid in terms of whether their contents were relevant and authentic with regard to what English second language speakers could reasonably be expected to know. Adult learners have differing English second language knowledge depending on the different contexts in which they have learned their second language. This thesis investigates the implications of the different contexts and different language needs of adult learners for the testing of English as a second language for placement purposes in ABET programmes. The thesis examined two placement tests to determine how the recognition of prior second language learning was being interpreted and how the interpretation affected the validity of the tests. Learners= perceptions of the assessment process and test content were elicited in order to determine whether a policy of transparency had been followed in the implementation of the assessment. It was also established what the goals of the organisation were in implementing an ABET programme. This research suggests that placement testing should be viewed holistically; in other words, the goals of the organisation and the level of transparency affect the validity of the placement test. The conclusions were that the placement tests were inauthentic since their contents excluded certain vital aspects of real life performance, namely, that related to the work context. The research revealed that if the placement testing process and the ABET programme are integrated into the culture of the organisation and if employees are remunerated when they have passed the different levels in the programme, the programme is likely to achieve a fair measure of success. Recommendations are that literacy should be viewed as based on a variety of contexts and uses and that therefore tests should be tailored to suit each particular organisation and should contain workrelated content. Furthermore, multiple methods of assessment should be considered.