In the first chapter, I develop a panel no-cointegration test which extends Pesaran, Shin and Smith (2001)'s bounds test to the panel framework by considering the individual regressions in a Seemingly Unrelated Regression (SUR) system. This allows to take into account unobserved common factors that contemporaneously affect all the units of the panel and provides, at the same time, unit-specific test statistics. Moreover, the approach is particularly suited when the number of individuals of the panel is small relatively to the number of time series observations. I develop the algorithm to implement the test and I use Monte Carlo simulation to analyze the properties of the test. The small sample properties of the test are remarkable, compared to its single equation counterpart. I illustrate the use of the test through a test of Purchasing Power Parity in a panel of EU15 countries. In the second chapter of my PhD thesis, I verify the Expectation Hypothesis of the Term Structure in the repurchasing agreements (repo) market with a new testing approach. I consider an "inexact" formulation of the EHTS, which models a time-varying component in the risk premia and I treat the interest rates as a non-stationary cointegrated system. The effect of the heteroskedasticity is controlled by means of testing procedures (bootstrap and heteroskedasticity correction) which are robust to variance and covariance shifts over time. I fi#nd that the long-run implications of EHTS are verified. A rolling window analysis clarifies that the EHTS is only rejected in periods of turbulence of #financial markets. The third chapter introduces the Stata command "bootrank" which implements the bootstrap likelihood ratio rank test algorithm developed by Cavaliere et al. (2012). The command is illustrated through an empirical application on the term structure of interest rates in the US.

This thesis collates a research programme of published papers completed by the candidate during the registration period of study, that are relevant to various aspects of feeding, eating and drinking difficulties (referred to as dysphagia) within a paediatric population from a speech and language therapy (SLT) perspective. Very few studies examine current SLT practice with this population. In the absence of research specific to the needs of children with congenital disorders, there are approaches being used by SLTs without a full rationale for their use and there is persistence in using therapy approaches that might not be beneficial for a child (Harding & Cockerill, 2014). Key themes present throughout this work include understanding the neurological and physiological underpinnings to an approach; being clear about a therapy rationale; creating therapy methods that consider the capacity of children who are neurologically and learning disabled and their caregivers and integrating communication more clearly into the management of dysphagia. The studies presented include: i) small case studies describing observations and analysis of communication during typical mealtimes; collaborative therapy programmes specifying strategies and sessions for use of Alternative and Augmentative communication (AAC); evaluation of a therapy programme to reduce aversion to tube feeding; ii) data on use of a straw to evaluate and record changes over age and gender; iii)evaluation of an SLT intervention to train staff within a special school ; iv)a pilot study and RCT investigating the use of non-nutritive sucking (NNS) to wean premature infants off tube feeding onto full oral feeding, and iv) case reports on use of NNS with infants with congenital disorders and the relationship between feeding difficulties and speech development. The studies presented contribute to the evidence base for SLT in a number of ways including describing and evaluating current practice and techniques through case studies and measuring the effectiveness of a SLT protocol through an RCT. The case studies highlight: the importance of checking the knowledge, skills and training of significant others in delivering therapy interventions, the importance of working collaboratively, specifying components of therapy programmes and time needed to implement them, and the importance of communication in its broadest sense within a meal time context. The non – nutritive sucking (NNS) RCT found that children in the intervention groups were able to leave hospital significantly sooner than in the control groups. However, unlike many other studies there was no difference in the time taken to be able to feed orally. There are a number of methodological issues to consider in evaluating the studies. The issues arising from conducting research within a complex clinical environment are discussed in Chapter 5. These complexities include using significant others to deliver therapy programmes, accurate understanding and descriptions of the premature population, the inclusion and exclusion criteria for studies, and consideration of infants with congenital and neurodevelopmental needs. The needs of and difference between parents are also considered. Evaluation of results also needs to take into account the paucity of tools to measure infants’ skills both by SLTs and parents and other professionals. Recommendations are made for future research. These include more studies to investigate accurate interpretation of infant states, improved descriptions and subsequent stratifications of infant participants; repetition of the RCT with larger sample and in range of settings, inclusion of follow up to 24 months with added measurements of feeding and language skills. The thesis papers also suggest more focus in future studies on the role of communication as a tool to manage risk within mealtimes.

Mild reading difficulties are a pervasive symptom in aphasia, but are little researched. Eye tracking research with neurologically healthy participants has demonstrated that reading is influenced by a number of information sources that are related to our experience with language. Two of these sources of influence, frequency and context, demonstrate that more probable words and structures are processed more quickly than those that are less probable. However, not much is known about probabilistic influences on the reading of people with aphasia at the sentence level. Two eye tracking experiments were conducted to establish whether or not frequency and context influence reading for people with aphasia in a way that is parallel to that of neurologically healthy participants. The first experiment examined the influence of word frequency and context on visual word recognition at the sentence level. The second experiment examined the influence of argument structure frequency on the reading of temporarily ambiguous sentences. Specifically, target sentences appeared after a context that cued a specific verb meaning and probabilistically associated argument structure. The target sentence was either consistent with or at odds with that context. The analysis of eye movements from both experiments revealed that people with aphasia have prolonged fixation durations and an increased proportion of regressions (backward fixations), indicative of their reading difficulties. Results from the first experiment demonstrated large effects of word frequency and context on both first pass and second pass eye movement measures by both groups. This suggests that frequency and context are related to both early and late processing stages of reading. However, differences were found between groups in a later processing stage where the aphasia group relied more on the context (top-down processing support) than the control group. The second experiment revealed that participants from both groups were sensitive to argument structure frequency when they read sentences that were temporarily ambiguous. Context cues facilitated the access of verb meaning and probabilistically associated argument structure, even though the individuals with aphasia showed delayed reading patterns. The context effect was found for both first pass and second pass eye movement measures, and was particularly strong for total fixation durations, which indicate re-reading behaviour. Overall, the outcome suggests the importance of considering multiple factors in sentence reading. Reading and sentence decoding by neurologically healthy individuals as well as individuals with aphasia are not only influenced by syntactic factors, but are also sensitive to factors relating to our language exposure. Results from the aphasia group are consistent with constraint-based theories of sentence comprehension and with slowed or reduced processing accounts.

This thesis uses cognitive linguistics and digital humanities techniques to analyse abstract conceptualization in a corpus of popular science texts. Combining techniques from Conceptual Integration Theory, corpus linguistics, data-mining, cognitive pragmatics and computational linguistics, it presents a unified approach to understanding cross-domain mappings in this area, and through case studies of key extracts, describes how concept integration in these texts operates. In more detail, Part I of the thesis describes and implements a comprehensive procedure for semantically analysing large bodies of text using the recently- completed database of the Historical Thesaurus of English. Using log-likelihood statistical measures and semantic annotation techniques on a 600,000 word corpus of abstract popular science, this part establishes both the existence and the extent of significant analogical content in the corpus. Part II then identifies samples which are particularly high in analogical content from the corpus, and proposes an adaptation of empirical and corpus methods to support and enhance conceptual integration (sometimes called conceptual blending) analyses, informed by Part I’s methodologies for the study of analogy on a wider scale. Finally, the thesis closes with a detailed analysis, using this methodology, of examples taken from the example corpus. This analysis illustrates those conclusions which can be drawn from such work, completing the methodological chain of reasoning from wide-scale corpora to narrow-focus semantics, and providing data about the nature of highly-abstract popular science as a genre. The thesis’ original contribution to knowledge is therefore twofold; while contributing to the understanding of the reification of abstractions in discourse, it also focuses on methodological enhancements to existing tools and approaches, aiming to contribute to the established tradition of both analytic and procedural work advancing the digital humanities in the area of language and discourse.

A growing phenomenon in British English is Regional Dialect Levelling. This is where accents lose their local characteristics in favour of more supralocal forms. The result is that different areas cease to have recognisably different dialects. For instance, neighbouring towns or villages become linguistically indistinguishable. Earlier elements of dialectal diversity are shaved off through processes of linguistic smoothing. This research focuses on two key issues: 1. The understanding of the mechanisms involved in regional dialect levelling; 2. How accounts of dialect levelling can inform models of sound change more generally. In this thesis I present an apparent time sociolinguistic study of regional dialect levelling in Hastings, a town on the coast of East Sussex, England. The study employs an empirical analysis of a number of ongoing sound changes. Specifically, the study examines three sound changes that, through previous analyses, have been shown to operate through different mechanisms: two features that are attributed to the externally motivated processes levelling and diffusion, and one internally motivated change driven by pressures inherent in the linguistic system. These contrasting mechanisms have been chosen in order to investigate a number of issues: first, to examine how each type of change may contribute to regional dialect levelling; and second, the analysis of these features enables a close examination of the interplay between external and internal forces of language change. More broadly, the evidence from this research is used to evaluate traditional principles of sound change in order to investigate how well they hold within a variety that is undergoing regional dialect levelling.

This thesis studies twelve word-initial phonaesthemes identified in the vocabulary of English. Phonaesthemes (Firth, 1930; 1935) are phonotactic sequences that recur in multiple words with similar meanings. While several scholars have discussed these phonaesthemes (e.g. Sturtevant, 1947; Marchand, 1966; Bowles, 1995), no study has yet provided a detailed quantitative account of the extent to which they are present in the vocabulary of English or any other language. Moreover, no study has yet investigated whether these patterns are associations that speakers actually perceive, or what might influence their perception. In response to these gaps in the literature, this thesis begins with a quantitative study of the extent to which the twelve phonaesthemes are present in the vocabularies of three languages; English, French and Polish. This involves calculating and comparing the number of different words that exhibit each phonaestheme in the three languages. The study then investigates whether native speakers of English, French and Polish actually perceive the phonaesthemes. This is measured by studying the extent to which speakers productively associate the twelve phonotactic sequences with their respective meanings in the context of coined words. Finally, the study investigates a number of factors that could affect the extent to which the phonaesthemes are perceptible patterns. The findings indicate clear quantitative evidence of all twelve phonaesthemes in the vocabulary of English. Nine of the phonaesthemes are also evident in the vocabularies of French and Polish; however, eleven of the twelve patterns are more pervasive in English. In addition, the productive experiments find that all twelve phonaesthemes are perceived by English speakers; though some are more consistently perceived than others. It is clear from these experiments that each phonaestheme is more widely and perceived in English than in French or Polish. Finally, it appears that one key factor affecting the perceptibility and thus productivity of the phonaesthemes is the number of different meanings with which the phonotactic sequences recur.

Research in translation studies has paid relatively little attention to translation pedagogy and how students, as potential professional translators, may understand and apply the range of current theories and concepts to practical translation tasks. This research explores the construction of translation knowledge and its application with groups of postgraduate students from two universities in China and one in the UK. In the light of new insights into translation teaching, this exploration focuses on the application of translation theories and pedagogical issues. The thesis investigates the three cases qualitatively to ascertain the extent to which the participants used theoretical elements, their perspectives on translation theory, together with comments from Chinese and British translation teachers and the documents of translation courses similar to the case study programmes. This study aims to provide empirically an in-depth understanding of the construction of translation knowledge and expertise related to students’ learning of translation, translation theories and their application and the complexity of these features in their particular social contexts of learning. The research employs the triangulated methods of analysing a translation task performed by students, subsequent interviews with them and with translation teachers, and the analysis of documents relating to British and Chinese translation courses. Data analyses within and across the three cases of translation programmes serve the exploration from triangulated perspectives. Additional complementary data are analysed to support the results. The findings indicate that the differences between and across the three groups are partly due to local contextual and institutional variation, but are largely due to different conceptions of the roles of translation theories and translation pedagogy. Explorations of the three programmes show the features between them, in particular how the two Chinese groups have inadequate knowledge of translation. The findings indicate how constraints on the two Chinese groups promote a tendency towards incompetence in translation performance, through relatively incoherent course content, lack of pedagogical planning, and lack of practical and professional training. This shows a need to reconceptualise translation pedagogy in the Chinese cases, and arguably in similar Chinese contexts; findings show that traditional language pedagogy with non-specialist teachers remains the chief element in translation pedagogy. The principal findings are further evaluated and interpreted by considering linguistic approaches, the use of translation briefs, the quality of translators, a holistic view of translation theories and the need for enhanced systematic translation pedagogy in Chinese translation studies contexts. A need is shown for Chinese students to learn to go beyond the mechanisms of putting one language in terms of another. Developments could include the holistic consideration of combining translation product-process-translator processes, as shown in a chart derived from Williams’ (2013). These findings are corroborated by the reflections and comments from university teachers of translation in China. The results contribute to a wider understanding of translation theories and pedagogy applicable to translation teaching in a Chinese context. Further suggestions include practical ideas to enhance translation pedagogy in China.

Considerable worldwide research has investigated incidental vocabulary learning from L2 reading, yet so far nothing has been published about the actual learning that comes from reading various texts. This study investigated incidental lexical growth and retention by Libyan university EFL majors who were involved in a two-month ER programme. Their vocabulary gain was measured 1 week after the participants completed the Extensive Reading (ER) programme, 2 weeks later and 9 months later. The value of this study is that it used an innovative approach, which was developed from a research design by Horst (2005). This included the electronic scanning of books and lexical frequency profiling, helping the researcher to create individualised corpus profiles from the entire set of different texts the participants read. This data was then used to select target words for each participant. The methodology was an experimental case study, which entailed an experimental and control group design. The participants were EFL learners who studied English as their subject of specialisation in one of eight Libyan state universities. An original number of 80 participants were randomly selected from the entire population in the English Language Department and assigned equally between the experimental and control groups. However, due to the fact that this study was carried out in very anomalous circumstances (during the Libyan uprising, which started on 15/02/2011), the number of participants who successfully completed the ER programme was affected (18 participants in each group). The study showed that by using a corpus analysis strategy, it was feasible to measure learners’ individualised pre-post treatment acquisition of the vocabulary they encountered in a large number of ER graded texts. The findings of the study demonstrate that ER significantly improved the Libyan EFL learners’ incidental vocabulary acquisition. By the end of the study, it was found that about the third of the target words had been acquired by the participants at both receptive and productive levels of knowledge. The findings further indicate that word repetition was an important factor for an incidental pick up of vocabulary from the ER. In relation to long-term retention rate of learning, the findings suggest that incidental word knowledge, acquired through ER, significantly declines over time.

The key purpose of this thesis was to study how vision and possible selves motivate second language (L2) learners in their learning, and the way non-linear and dynamic patterns are exhibited in motivated L2 behaviours. The thesis consists of four studies that investigated the motivation of different target populations (secondary and university learners, and Christian Language Professionals) by using a variety of research methodologies, including retrodictive qualitative modelling (RQM), mixed methods approach, qualitative in-depth interviews and intervention. In Studies I and IV, the motivational trajectories of secondary school students and Christian Language Professionals were explored through the lens of dynamic systems theory (DST). In Study I, using retrodictive qualitative modelling (RQM), in which the conventional research data collection methodology was reversed, the results showed that seven learner archetypes exist in the teachers’ minds, and different unique motivational trajectories known as signature dynamics (with different combinations of attractor states) are present in the learners’ systems. The strength and weaknesses in applying the RQM model were identified. Study IV explored the developments of vision in four CLP through in-depth interviews and three different vision integration patterns were revealed: ‘fully integrated as a person’, ‘fully integrated Ideal CLP Self’, and ‘partially integrated CLP Self’. Study II examined the interrelationships among various sensory capacities (visual and auditory), imagery capacities, future self-guides and criterion measures in two target languages (English and Mandarin) in 172 Year-8 secondary school students in Hong Kong. Using both questionnaires and post-survey focus group interviews, it was found that learners’ future self-guides are endowed with visual, auditory and imagery capacities as the main components. Study III investigated the effects of an imagery-training intervention on the quantitative and qualitative change of possible L2 self-guides in 80 second-year science university students. A significant increase in the Ideal L2 Self and a significant decrease in the Feared L2 Self were shown, which suggested that the intervention has positive effects on learners’ future self-guides. In summary, the results from the collection of studies showed the different characteristics, effects, and motivational forces of vision and possible selves in the complex world of L2 learning and teaching. Along with the insights of the complex dynamic interplay between the factors in L2 learning and teaching, I have built a strong case for vision and possible selves as key motivational tools in the L2 classroom.

This thesis considers what I call lexical planning initiatives for Welsh – formal attempts to codify and standardise Welsh words. Welsh has been subjected to lexical planning – and purification – attempts for a number of centuries, with lexicographers seeking to coin and standardise Welsh-equivalent words for concepts that have initially emerged through contact with English (and other prestigious languages). Lexical planners have redoubled their efforts in the last fifty years, but especially since 1993, largely as a result of the language revitalisation movement. Lexical planning efforts can be envisaged as attempts to influence the acquisition and use of any lexical resources, but they often focus on specific subject matters, especially from modern or emergent domains or disciplines. Such initiatives are often referred to as terminology planning/standardisation. My research considers the implementation of these planning initiatives, focusing on spoken language data at two research sites: the broadcast media and an office-based workplace. Taking a two-pronged approach to analysis, I ask whether, how and why Welsh speakers use planned lexis. First, I consider the extent to which the lexical content is in keeping with the stipulations of lexical planners in their codification texts. This approach is chiefly quantitative, drawing broadly on corpus linguistics and variationist sociolinguistics. Secondly, taking a more context- and practice-focused, as well as critical approach, I undertake an interaction analysis of the in situ use of lexical resources. From this perspective, we gain a picture of the underlying, sometimes conflicting, ideologies and discourse priorities that motivate lexical choice. This approach considers lexical planning initiatives not as implemented top-down, but embedded in their social milieu. Finally, I consider the implications of my research for the broader revitalisation effort, asking to what extent lexical planning initiatives, as they are currently imagined and conducted, complement other language planning endeavours, and whether and how they might be reconsidered.