The fieldwork component of this study comprised of academic English language tests with 120 participants and 30 open-ended in-depth interviews with first year undergraduate university students in Fiji. To this end, the fieldwork involved administering academic English language tests, using writing interventions and using these to evaluate educational inequalities faced by the students. This process was aided by the use of open-ended questions. The participants were required to sit two academic English language writing tests, one at the beginning of their first year and one at the end of the first year. This research was carried out as a longitudinal study by administrating a writing test in the second week of the first year (beginning) of their university program, followed by a second test at the end of their first year, namely, in the final week of classes in semester two of the year. The test was conducted at the beginning and at the end of their first year which lasted 1 hour. There were three writing interventions and feedback was given throughout the yearlong study. The writing interventions were academic essays, paragraph writing and summary writing. Tasks in the writing intervention involved students to write and submit to the researcher in their leisure time. I provided feedback on each of the three interventions individually to the cohort after assessing them throughout the year. Feedback involved highlighting nonstandard forms of writing style or grammar, discussing ways of improving the writing pieces and suggesting resources on academic writing. A total of 30 interviews (30 - 40 minutes each) were conducted at the end of the participants' first year via Zoom and on Skype. Volunteers from the same cohort of 120 participants were recruited at random based on their performance in the tests, both high performers as well as low performers were interviewed. The interviews were conducted after the end of the students' one-year university program.

This project investigates the nature and extent of educational inequalities or injustices that manifest in academic language testing evaluation regimes. The focus is on Fiji, a multilingual, multicultural country situated in the South Pacific. The major languages spoken in the country are Fijian (iTaukei), English, and Fiji Hindi. Other languages spoken in the community include Gujarati, Punjabi, Urdu, Telugu, Tamil, Mandarin, and Cantonese. Due to the colonial history of Fiji, English has emerged as the main lingua franca among the various ethnolinguistic groups in the country. English is also the sole medium of instruction at all three universities in the country. The dominance of English in the context of tertiary education results in a sociocultural problem when students enter universities in Fiji from a variety of high school backgrounds. Some come from urban or semi-urban high schools, while others come from rural high schools, which are located in the interior of the country and on smaller islands. Depending on the school, students may acquire different levels of English proficiency, which may leave them ill-equipped for university study. This situation directly draws attention to educational inequalities that students face at the tertiary level due to the variable levels of English language skills provided to them at the primary and secondary levels.

Drawing on the review of the relevant literature and ethnographic research, this study uses grounded theory methodologies to gain insight into the underlying educational inequalities and injustices emanating from lack of epistemic access for those university students with inadequate English language skills. To this end, the study analyses the level of difference in writing abilities at the beginning and end of the first-year of study. The findings and conclusions from this study hold key insights for informing ongoing efforts towards ameliorating educational inequalities in Fiji. The study was conducted at a university in Fiji. The first-year enrolment number was 400 students. One hundred and twenty students (30% of the student cohort) were sampled at the beginning of the first year and at the end of their first year of university program. The same cohort of 120 students was tracked throughout the project, out of which 30 students were interviewed at the end of the first year to assess the writing interventions and to gauge the nature and extent of educational inequalities faced by the participants. The interviews were analysed using grounded theory methodology to generate themes from raw data.

The study uses the Common European Framework of Reference for Languages (CEFR) to evaluate student writing skills. The CEFR is one of the most comprehensive frameworks for language evaluation prevalent today, and has been widely used by language testing organisations in Western countries. However, since its introduction in Europe in 2001, the CEFR has so far not been applied to the unique sociolinguistic context of the South Pacific. This study is the first to do so, and gauges the framework's relevance and usefulness in a non-European multilingual context.

The outcomes of this research make significant contributions in informing policy discourse on reversing educational inequalities at the tertiary level. Firstly, it reaffirms the CEFR's capability in successfully gauging students' preparedness for English-language university study and work life. Secondly, it attests to the CEFR's suitability for use in the intensely multilingual (or plurilingual) South Pacific context. Most importantly, the study identifies four specific indicators of educational inequalities that Fijian students in higher education encounter from primary school through to high school and university that hinder the development of adequate academic English language skills. These include: (i) lack of teaching and learning resources, (ii) language barriers, (iii) problems with the medium of instruction, and (iv) shortage of experienced teachers. Although these indicators of educational inequalities are reported in comparable studies from other parts of the world, they are significant findings for this study, not least because the South Pacific is a geographical locale where individual nation-states such as Fiji consist of several islands with unique socio-cultural formations and complex linguistic landscapes.

The methodological contributions and the unique data set of the study will advance scholarly and social policy conversations on this topic. Through analysing the rich narrative stories of research participants and qualitative data, the study makes an original contribution to the body of knowledge on how grounded theory research methodologies can be applied to a longitudinal language testing research context. At present, language testing in higher education relies on data from conventional formative and normative assessments. Approaches such as grounded theory and longitudinal research design have rarely been used in intensely multilingual contexts such as the South Pacific. The findings of this study hold promise in informing higher education policies for improving support systems to enhance the smooth transition of multilingual students from high school through to university and into the workforce.

This thesis is on the sociolinguistics of writing. It investigates the use of scripts for the Sindhī language of South Asia, both from a diachronic and synchronic perspective. The thesis first analyses the rich but understudied script history of the Sindhī language from the tenth century to modern times. In doing so, it investigates the domains in which certain scripts were used, and identifies definite patterns in their distribution. Particular attention is paid to Perso-Arabic and Devanāgarī, which emerged as the two most widely used scripts for the language in the twentieth century. The diachronic analysis draws on several linguistic, literary and other academic works on the Sindhī language and brings to the fore hitherto neglected data on historical script use for the language. The thesis then presents and analyses oral interview data on community opinion on the recent proposal to use the Roman script to read and write Sindhī. The synchronic analysis is based on original fieldwork data, comprising in-depth qualitative interviews with fifty members of the Indian Sindhī community of diverse backgrounds and ages from various geographical locations. Empirically, this work is one of the first to provide a comprehensive diachronic and synchronic review and analysis of script practices in the Sindhī community specifically from a sociolinguistic perspective. It also provides revealing insights into the kinds of expectations an urbanised, highly educated and socioeconomically successful minority has of a writing system for its language. In doing so, the study challenges the prevalent simplistic claim in the literature that minority communities are desirous of seeing their language in writing. Most importantly, this work indicates the emergence of a so-called new variety of Sindhī phonology in India, which differs subtly from the old variety phonology. The implications of this subtle shift in phonology for Sindhī pedagogical material form a key part of the findings of this study. Theoretically, this work contributes to the concept of orthographic transfer, which is the phenomenon of phoneme-grapheme correspondences in a particular orthography being inadvertently applied to another orthography. The study also affirms the presence of a scriptal diglossia, or digraphia, in script use for the Sindhī language, where the use of particular scripts for the language is implicitly determined by domain and context. The potential impact of orthographic transfer and digraphia on the pedagogy of lesser-learnt languages is a key part of the study’s findings. Methodologically, the juxtaposition of historical and present-day sociolinguistic factors at play offers a fresh and nuanced look at the rise and fall of scripts in the context of a language with a centuries-old written tradition. The study concludes that usage of a particular script for a language is not the result of a simplistic binary opposition between authoritarian imposition and voluntary choice. Rather, it is a reflection of several pragmatic and symbolic considerations by the community in question. The thesis puts into perspective the various psychological, socioeconomic and cultural forces at work in determining script use for the Sindhī language. In doing so, the thesis makes several additions not just to the existing body of knowledge on the Sindhī language, but also to the fledgling field of inquiry that is the sociolinguistics of writing. These varied and unique contributions set the study apart from previous research on the subject.