SEU Approaching the Research Design Paper

User Generated

Tuvqnqn

Writing

Saudi electronic university

Description

Consider the following required readings and resources, which provide information on how to work out the design portion of the research. Specifically: 

Guidelines for the Preparation of your Master’s Thesis (Required reading)

IMRD: The Parts of a Research Paper (video in the Module lecture)

  1. Chapter 7 in Research Methodologies in Translation Studies (Required reading)

After completing the readings and watching the video, return to the research project on translation and decide which structure and method you think would be appropriate for your project.

Selecting strategic components and elements in the design structure is key for each specific type of inquiry you want to conduct research on. 

  1. Use the guidelines in the generic references in the references cited in points (1 &, 2) and then apply them to the translation-specific guidelines cited in point (3). 

Justify why you chose the quantitative or qualitative research method and structure.

If you decide to use both quantitative and qualitative methods (the mixed approach method), please explain why you decided to use both of them in your reporting.

Unformatted Attachment Preview

Research Methodologies in Translation Studies Gabriela Saldanha and Sharon O’Brien First published 2013 by St Jerome Publishing Published 2014 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN 711 Third Avenue, New York, NY, 10017, USA Routledge is an imprint of the Taylor & Francis Group, an informa business Copyright © Gabriela Saldanha and Sharon O’Brien 2013 Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. British Library Cataloguing in Publication Data A catalogue record of this book is available from the British Library Library of Congress Cataloging-in-Publication Data Saldanha, Gabriela, author. Research methodologies in translation studies / Gabriela Saldanha and Sharon O’Brien. pages cm Includes bibliographical references and index. ISBN 978-1-909485-00-6 (pbk : alk. paper) 1. Translating and interpreting--Research--Methodology. I. O’Brien, Sharon, 1969- author. II. Title. P306.S244 2013 418’.02--dc23 2013030989 ISBN: 978-1-909485-00-6 (pbk) Delta Typesetters, Cairo, Egypt For Fionn, Rebeca, Rián, Martina This page intentionally left blank Research Methodologies in Translation Studies As an interdisciplinary area of research, translation studies attracts students and scholars with a wide range of backgrounds, who then need to face the challenge of accounting for a complex object of enquiry that does not adapt itself well to traditional methods in other fields of investigation. This book addresses the needs of such scholars – whether they are students doing research at postgraduate level or more experienced researchers who want to familiarize themselves with methods outside their current field of expertise. The book promotes a discerning and critical approach to scholarly investigation by providing the reader not only with the know-how but also with new insights into how new questions can be fruitfully explored through the coherent integration of different methods of research. Understanding core principles of reliability, validity and ethics is essential for any researcher no matter what methodology they adopt, and a whole chapter is therefore devoted to these issues. While necessarily partial, the survey presented here focuses on methodologies that have been more frequently applied and therefore more thoroughly tested. It is divided into four different chapters, according to whether the research focuses on the translation product, the process of translation, the participants involved or the context in which translation takes place. An introductory chapter discusses issues of reliability, credibility, validity and ethics. The impact of our research depends not only on its quality but also on successful dissemination, and the final chapter therefore deals with what is also generally the final stage of the research process: producing a research report. Gabriela Saldanha is a Lecturer in Translation Studies at the Department of English Language and Applied Linguistics, University of Birmingham, UK, where she convenes both the distance and campus-based MA programmes in Translation Studies. Her research has focused on gender-related stylistic features in translation and on translator style, using corpus linguistics as a methodology. Her teaching focuses on translation theory, research methods and translation technology. She is co-editor of the second, revised edition of the Routledge Encyclopedia of Translation Studies (2009). She is co-editor of Translation Studies Abstracts and is on the editorial board of InTRAlinea. Sharon O’Brien is a Senior Lecturer in Translation Studies at the School of Applied Language and Intercultural Studies, Dublin City University, Ireland, where she teaches postgraduate and undergraduate courses in Translation Studies. Her research has focused on translation technology, especially the post-editing of machine translation output, translation processes, and controlled authoring using keyboard logging, screen recording and eye tracking. Her teaching focuses on translation technology, software localization, translation theory and research methods. She is co-editor of St. Jerome’s Translation Practices Explained series and a track editor for the journal Translation Spaces. This page intentionally left blank Contents Acknowledgements xiii Chapter 1. Introduction 1.1 Motivation and Intended audience 1.2 Scope and limitations 1.3 Research model, structure and content of the book 1 2 5 Chapter 2. Principles and ethics in research 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 Introduction Ontology and epistemology Research terminology Types of research Research questions and hypotheses The literature review Data Qualitative, quantitative and mixed-methods approaches Research operationalization 2.9.1 Measurable variables 2.10 Research quality 2.10.1 Validity 2.10.2 Reliability 2.10.3 Generalizability 2.10.4 Qualitative methods, credibility and warrantability 2.11 Research ethics 2.11.1 Ethics in context 2.11.2 Ethics approval 2.11.3 Informed consent 2.11.4 Deception 2.11.5 Power relations 2.11.6 Protection from harm 2.11.7 Internet-mediated research 2.11.8 Plagiarism 2.12 Summary 10 10 12 14 16 19 20 22 23 25 27 27 35 36 38 41 42 43 43 45 45 46 47 48 49 Chapter 3. Product-oriented research 3.1 Introduction 3.2 A descriptive/explanatory approach to the analysis of language 50 50 3.3 3.4 3.5 3.6 3.7 3.2.1 Critical discourse analysis 3.2.2 Corpus linguistics 3.2.3 Strength and weaknesses of critical discourse analysis and corpus linguistics Designing studies in critical discourse analysis and corpus linguistics 3.3.1 Corpus-driven, corpus-based, argument-centred and problem-based designs 3.3.2 Selecting and delimiting texts as units of investigation 3.3.3 The need for comparative data 3.3.4 Corpus typology Building corpora 3.4.1 Corpus design criteria 3.4.2 Annotation and alignment Analyzing texts and corpora 3.5.1 The linguistic toolkit 3.5.2 Fairclough’s relational approach to critical discourse analysis 3.5.3 The tools of corpus analysis 3.5.4 Addressing issues of quality in critical discourse analysis and corpus linguistics Research on translation quality assessment – Introduction 3.6.1 Strengths and weaknesses 3.6.2 Design 3.6.3 Which QA model(s)? 3.6.4 Data collection 3.6.5 Analysis Summary 51 55 57 61 61 64 66 67 70 71 76 80 80 83 85 92 95 96 97 100 105 107 108 Chapter 4. Process-oriented research 4.1 Introduction 4.1.1 Common topics 4.2 General translation process research issues 4.2.1 Design 4.2.2 Data elicitation 4.2.3 Analysis 4.3 Introspection 4.3.1 Design 4.3.2 Data elicitation 4.3.3 Transcription 4.3.4 Analysis 4.4 Keystroke logging 4.4.1 Design 4.4.2 Data elicitation 4.4.3 Analysis 109 111 113 113 118 119 122 124 126 128 130 132 133 134 135 4.5 Eye tracking 4.5.1 Design 4.5.2 Data elicitation 4.5.3 Analysis 4.5.3.1 Analysis of temporal data 4.5.3.2 Analysis of attentional data 4.5.3.3 Analysis of data pertaining to cognitive effort 4.5.3.4 Analysis of linked data 4.6 Complementary methods 4.6.1 Contextual inquiry 4.6.2 Personality profiling 4.6.3 Physiological measurements 4.7 Summary 136 138 141 142 143 143 144 145 145 145 146 148 148 Chapter 5. Participant-oriented research 5.1 Introduction 5.2 Questionnaires 5.2.1 Overview 5.2.2 Strengths and weaknesses 5.3 Designing questionnaire surveys 5.3.1 Operationalization 5.3.2 Number and phrasing of questions 5.3.3. Open and closed questions 5.3.4 Likert scales 5.3.5 Pilot testing 5.3.6 Reliability and validity 5.3.7 Ethical considerations 5.4 Data collection using questionnaires 5.4.1 Sampling 5.4.2 Response rate 5.4.3 Internet-mediated collection methods 5.5 Interviews and focus groups 5.5.1 Overview 5.5.2 Strengths and weaknesses 5.6 Designing interviews and focus groups 5.6.1 Types of interviews and focus groups 5.6.2 Designing interview and focus group schedules 5.6.3 Language issues 5.6.4 Piloting 5.6.5 Ethical considerations 5.7 Eliciting data using interviews and focus groups 5.7.1 Sampling and recruiting participants 5.7.2 Interviewing and moderating: Basic principles and key challenges 150 151 151 152 153 153 154 157 157 158 159 161 163 164 165 166 168 168 169 171 172 174 177 178 179 180 180 184 5.7.3 Face-to-face, telephone and Internet-mediated interviews and focus groups 5.8 Analyzing qualitative data 5.9 Analyzing quantitative data 5.10 Data analysis in mixed methods research 5.11 Summary 186 188 194 201 204 Chapter 6: Context-oriented research: case studies 6.1 6.2 6.3 6.4 Introduction Definition of case study When to use case studies Case study design 6.4.1. Types of case studies 6.4.2 Delimiting case studies 6.5 Collecting data 6.5.1 Written sources 6.5.2 Verbal reports 6.5.3 Observation 6.5.4 Physical artefacts 6.5.5 Quantitative data 6.5.6 Using a database to manage data 6.5.7 Ethical considerations 6.6 Analyzing case-study data 6.6.1 General principles 6.6.2 Practical suggestions 6.6.3 Computer-aided qualitative analysis 6.7 Summary 205 207 209 211 211 215 217 218 220 221 224 224 224 225 227 227 230 231 232 Chapter 7: Conclusion: The research report 7.1 Introduction 7.2 Structuring the report 7.3 Framing the report: introduction, literature review and conclusion 7.4 Reporting methods 7.5 Reporting qualitative data 7.6 Reporting quantitative data 7.7 Reporting linguistic data 7.8 Summary 234 234 References 244 Index 270 236 237 240 241 242 242 List of Figures Figure 2.1 Example of Research Terminology Applied to TS Figure 2.2 A theoretical model expressing relationship between time pressure (X axis) and translation quality (nominal scale on the Y axis) 14 Figure 3.1 Design of a bidirectional parallel corpus Figure 3.2 Sample header file from the Translational English Corpus Figure 3.3 Example of tagged text taken from the British National Corpus Figure 3.4 Example of a parallel concordance obtained with Paraconc Figure 3.5 Fairclough’s relational model (adapted from Fairclough 2003:36) Figure 3.6 A concordance of the node ‘source’ obtained using the BYU_BNC interface for the British National Corpus Figure 3.7 A sketch engine profile of the word ‘source’ 68 77 Figure 4.1 Example of TAP transcription from the TransComp project (http://gams.uni-graz.at/fedora/get/o:tc-095-201/bdef: TEI/get), post-phase_2, participant: Professional AEF) Source Text A1 Figure 4.2 Example of Translog linear data Figure 4.3 Example of fixations during a reading task from a study by Doherty and O’Brien (2012) Figure 5.1 Example of bell curve in normal distribution 26 78 79 84 90 91 129 132 137 197 List of Tables Table 2.1 Threats to validity identified by Frey et al. (1999) 30 Table 5.1 Examples of coding of semi-structured interview data Table 5.2 Example of quartile data: from processing times measurement in Guerberof (2008:38) 191 198 This page intentionally left blank Acknowledgements Many people assisted us in the writing of this book and we are enormously grateful for that assistance, no matter how small or large the contribution. We would especially like to thank Jenny Williams for providing us with insightful and helpful feedback on a draft of this book. We drew on the expertise of many others for advice and feedback on specific chapters and we express our sincere gratitude to them: Fabio Alves, Andrew Chesterman, Claire Hewson, Dorothy Kenny, Kaisa Koskinen, Ian Mason, Rebecca Tipton. Any errors are, of course, of our own making. Finally, we are hugely grateful to our families for their patience and support. This page intentionally left blank Chapter 1. Introduction 1.1 Motivation and Intended audience Recent years have witnessed an increase in the number of translation training programmes across the world, with a resulting explosion in the number of masters and doctoral students and, as reported in Mason (2009a), a concomitant move towards explicit forms of research training in translation studies. The book entitled The Map: A Beginner’s Guide to Doing Research in Translation Studies, co-authored by Jenny Williams and Andrew Chesterman and published in 2002, was given a very warm welcome by the translation studies community and is still highly regarded by established and novice researchers alike. Clearly there was, and still is, a thirst for a book that was specifically focused on research within the domain of translation studies. Since the publication of The Map, there have been some methodological developments in the field with, for example, the application of methods such as keystroke logging, eye tracking, Internet-mediated research, as well as an increased focus on sociological and ethnographic approaches to research and on research ethics. We feel it is now time to build on the excellent foundation set by Williams and Chesterman. The Map establishes the foundations of translation studies research and is particularly useful for those who are starting to think about doing research in this area, and who need to decide between different areas of research, theoretical models, types of research, and so on. The focus of this book is on specific methodologies. We describe in detail when and how to apply different methodologies and we provide examples from translation studies research. There are, already, many excellent publications that describe how these methodogies are applied in related domains such as applied linguistics, social science, psychology and cultural studies. These books are, of course, valuable to the translation researcher. However, it is our experience that even in related disciplines the books fail to answer all our questions about doing research in translation. Often the examples feel distant or even irrelevant, thus failing to inspire translation studies researchers. We are convinced that discussing methodologies within the translation studies context and offering examples of current best practice has a value above and beyond standard, generic textbooks. The Map is a beginner’s guide, as stated in the title, and is mostly directed at PhD students. This book will also hopefully be useful to PhD, Masters and Undergraduate students. Research students are expected to develop core research skills, such as understanding what counts as creativity, originality, and the exercise of academic judgement. We have kept these needs in mind during the writing process. However, we feel that a need exists beyond this readership too. As discussed below, translation studies is interdisciplinary by nature. While the professionalization of translation and the recognition of translation as an academic discipline have resulted in translation-specific educational pathways all the way from the undergraduate to the doctoral level, the field of translation  Gabriela Saldanha and Sharon O’Brien studies continues to attract researchers from many different backgrounds who may not be familiar with the wide range of methodological practices in the field. By bringing together in one publication methodologies originating in different disciplines and discussing how they can be fruitfully combined for the study of translation we aim to contribute to the cross-fertilization of the different research practices that inform translation studies. 1.2 Scope and limitations Guba and Lincoln (2005:191) argue that “[m]ethodology is inevitably interwoven with and emerges from the nature of particular disciplines”. Linguistics and literary criticism were for a long time the main source of theories and methods in translation research, which was based on comparative text analysis carried out with varying levels of linguistic or literary insight. Much of the research on literary translation is still embedded within a comparative literature framework and linguistic approaches are still widely used, although rarely with the same narrow focus they initially adopted. During the 1980s, translation scholars began to draw more heavily on methodologies borrowed from other disciplines, including psychology, communication theory, anthropology, philosophy and cultural studies (Baker 1998:278). More recently, the importation of theories and models from social theory has again widened the range of methodologies applied within translation studies. In 1998, Baker suggested that: Although some scholars see translation studies as interdisciplinary by nature (Snell-Hornby 1988), this does not mean that the discipline is not developing or cannot develop a coherent research methodology of its own. Indeed, the various methodologies and theoretical frameworks borrowed from different disciplines are increasingly being adapted and reassessed to meet the specific needs of translation scholars. (Baker 1998:279) The picture emerging from the current book is not of a single coherent methodology that could be described as specific to translation studies. However, there is indeed evidence of adaptation and reassessment, which is perhaps most clear in the dynamism with which different theoretical frameworks and methodologies are being combined for the purpose of addressing translation studies’ concerns, since none of the methodological apparatuses of related disciplines can, on their own, fully account for translation phenomena (see, for example, the 2013 special issue of Target on interdisciplinarity in translation process research). In their overview of the main paradigms in contemporary qualitative research, Guba and Lincoln (2005:191) note that “[i]ndeed, the various paradigms are beginning to ‘interbreed’ such that two theorists previously thought to be in irreconcilable conflict may now appear, under a different theoretical rubric, to be informing one another’s arguments”. We believe that translation studies has now successfully moved beyond the paradigm conflicts of the 1990s (Baker Introduction  1998) and has succeeded not only in celebrating and accepting a diversity of approaches but in ‘interbreeding’ for the benefit of a more comprehensive and nuanced account of the object of study. In terms of this particular book, while we have made an effort to reflect the interdisciplinarity of translation studies and have tried our best to represent the different epistemological and ontological positions within the discipline, our account is necessarily partial, reflecting our own academic backgrounds and experiences. We have aimed to remain aware, insofar as our in-built biases allow us, that our way of seeing and thinking about research methods may not necessarily be in agreement with the way others see or think about them. In what follows we justify our choices and acknowledge their limitations regarding the contents of the book and the views they reflect. Translation studies is interdisciplinary not only because it borrows from a wide range of disciplines but also because it covers a wide range of practices. While we have made an attempt to reflect this diversity in the examples we have selected for discussion, there are areas of translation research that are not adequately covered by the methodologies described here, such as interpreting and translation history. In her reflections on the periods of fragmentation experienced by translation studies while fighting to establish itself as an academic discipline in its own right, Baker (1998:279) mentions the fact that theoretical models in translation studies have tended to ignore interpreting and produced research that is of no relevance to those interested in that field. While our impression is that there has been progress in this regard, our experience is mainly within translation studies and we may not be the best people to judge. We see interpreting studies at the forefront of many of the methodological advances in the discipline in recent years, and this view is reflected here in recurrent examples from interpreting studies, particularly in the discussion of critical discourse analysis, interviews and focus groups. However, we also acknowledge that the nature of interpreting as spoken interaction presents certain challenges in terms of research methodology which we are not in a position to discuss in detail. The same could be said about translation history. The methodology described in Chapter 6, case studies, has often been used in historical translation research and two of the examples used in that chapter deal with historical phenomena. However, the specificities of researching the past are not the focus of the chapter. It is worth noting that translation history is the one area of translation studies research where there is a book-length publication devoted to questions of methodology (Pym 1998). One topic that has dominated the literature in translation studies in the past few years is the question of centre and periphery, dominant and subservient, Western and non-Western perspectives, and we feel it is important to reflect on these matters in relation to our approach. A question we have often asked ourselves while writing this book is: how ‘universal’ are the research methods described here? Susam-Sarajeva (2002) helpfully rules out the use of the terms ‘Western/non-Western’. She argues that “[b]eing ‘non-Western’ has apparently  Gabriela Saldanha and Sharon O’Brien become the only common denominator behind otherwise vastly different languages and cultures” (ibid.:193). Equally, “the same dichotomy renders ‘the West’ more homogeneous than it actually is” (ibid.). She argues instead for the terms ‘centre’ and ‘periphery’ but acknowledges that these are also problematic. Susam-Sarajeva highlights the danger that those operating in ‘the periphery’ will regard their own concepts and ways of thinking as inferior; they will be “’educated away’ from their own culture and society” (ibid.:199). This, she says, is inevitable, because for research to be considered ‘useful’, ‘publishable’ and ‘quotable’ it must refer to the established (central) frameworks. In order to be rated highly as a researcher, one needs to publish in specific journals, most of which use English as the language of publication and prioritize their own research agendas, with their concomitant limitations in terms of research models and methodologies. One of the authors of this book originates from South America and left behind a country and a language to pursue an academic career. The other originates from and lives in a former colony on the periphery of Europe. Therefore, these issues are close to our hearts as individuals. Nevertheless, there is no denying that our academic perspective is ‘central’ in Susam-Sarajeva’s terms, even if this is by (de)fault rather than choice, since it reflects the environment in which we have been immersed during our academic careers and the one in which we operate more comfortably. We can think of many good reasons for academics to start operating outside their comfort zones, but we take the view that a book on research methodologies is not the best place to do that. However, in the writing of this book we do not intend to present specific frameworks as the only relevant ones and we hope that they are relevant as one way to do things no matter where the research is being conducted or where the researcher or the researched come from. Our expertise is limited mainly to empirical research, which not only has implications for the scope of the book, focusing on empirical methods and methodologies, but probably also permeates the content in a more pervasive and subtle manner in terms of our assumptions as to what constitutes good academic practice, with its emphasis on evidence, hypotheses and operationalization. Despite acknowledging our limited focus, we have chosen not to call the book ‘empirical research methodologies’ because we do not believe in a clear-cut distinction between conceptual and empirical research. Good empirical research needs to be based on conceptual research and conceptual research, to be useful, needs to be supplemented by evidence. Evidence and theory are crucial to all researchers: “[y]ou need the ‘facts’ – imperfect though they may be; and you need to be able to understand or explain them (theory)” (Gillham 2000:12). Although we generally talk about theories as the basis on which we build our empirical studies, we should not forget that theory is also what researchers create, the way they account for the data, particularly in inductive approaches to research (see Chapter 2). In other words, research can be seen as theory building as well as theory testing; as providing answers (for example, in hypothesis-testing research) as well as framing questions (in hypothesis-generating research). A further clarification to be made in relation to our understanding of empirical Introduction  research is that we do not believe that empirical research is necessarily descriptive or incompatible with critical-interpretive approaches. There has been a tendency in translation studies to equate empiricism with descriptivism and the latter with a-historical and uncritical methods that aim to produce generalizations about patterns of translational behaviour with predictive power (Crisafulli 2002). While there is a need for non-prescriptive research that establishes what translators normally do and why, as opposed to telling translators what to do, we also agree with Crisafulli that this does not mean that description must, or can, be non-evaluative: “value judgements influence the selection of data as well as the descriptive categories of analysis and the explanatory theories into which these are organized” (2002:32). When describing a phenomenon we inevitably foreground certain relationships at the expense of others and thus prioritize certain explanations over others. For example, in a corpus-based study of explicitation, the design of the corpus (whether it is comparable or parallel, whether it includes translations from more than one language or from more than one translator) will necessarily limit the otherwise extremely wide range of factors that could be seen as having an impact on the frequency of instances of explicitation to be found in the corpus. A selfreflective approach to research should acknowledge this inherent bias while at the same time highlighting the benefits of exploring certain variables in depth at the expense of excluding others. It should also look for potentially contradictory evidence as well as seek to back up any results with relevant data from other sources. We consider methodological triangulation to be the backbone of solid, high quality research and so it is implicitly suggested throughout each chapter. 1.3 Research model, structure and content of the book Empirical research involves gathering observations (in naturalistic or experimental settings) about the world of our experience. Generally, the choice of what aspect to observe will impose certain restrictions in terms of the methods we use. Therefore, we have chosen to divide the chapters of this book according to the focus of our observations: the texts that are the product of translation (Chapter 3), the translation process (Chapter 4), the participants involved in that process (Chapter 5) and the context in which translations are produced and received (Chapter 6). It is important to stress, however, that (1) whether a piece of research is process-, product-, participant- or context-oriented is not determined by the methodology itself or even the source of data but by the ultimate aims of the researcher, and (2) when investigating any of these aspects of translation it is impossible to exclude from view all the others; there is inevitable overlap. We are aware that, in adopting this division of translation phenomena, we are offering the outline of yet another model of translation studies research, rather than drawing on those already proposed by, for example, Marco (2009) or Chesterman (2000). Our model of translation research is by no means flawless or complete; it reflects the perspectives from which translation has been viewed rather than those from which we could possibly view it. In what follows  Gabriela Saldanha and Sharon O’Brien we explain how this model compares to Chesterman’s and Marco’s. Chesterman distinguishes three types of models: comparative models, which aim to discover language-pair translation rules, language-system contrasts, or translation product universals (also known as features of translation); process models, which represent change (from state A to state B) over a time interval (although the process is not necessarily linear) and allow us to understand decision-making in translation and cognitive factors influencing this process; and causal models, which aim to explain why translations are the way they are by reference to three dimensions of causation: the translator’s cognition (translator’s knowledge, attitude, identity, skills), the translation event (translator’s brief, payment, deadlines) and the socio-cultural factors (ideology, censorship, cultural traditions, audience). Chesterman (2000:21) argues that the causal model is “the richest and most powerful” because it contains the other two models – linguistic and cognitive factors are taken as causal conditions, and effects at the linguistic and cognitive levels are recorded – and it encourages explanatory and predictive hypotheses. On a superficial level, we could say that the product-oriented methodologies described in Chapter 3 correspond to Chesterman’s comparative model; the process-oriented methodologies described in Chapter 4 to the cognitive one; the context-oriented methodologies described in Chapter 6 to the causal one; and participant-oriented methodologies might be mapped either onto a cognitive or causal model according to the precise focus of the research. However, as explained above, an underlying assumption of our approach is that there cannot be purely descriptive (comparative or procedural) research, because any (good) research necessarily takes into account possible explanations, and descriptions are never neutral. Therefore, another way of mapping our model onto Chesterman’s three types would be to classify it in its entirety as a causal model that recognizes three different dimensions of causality (linguistic, cognitive and contextual). A rather complex issue that is necessarily brought to the fore by this mapping of models and which cannot be addressed in much detail here is the potentially different ways of understanding causality in the two approaches (Chesterman’s, as outlined in his 2000 publication, and ours). Our understanding is very broad: we simply suggest that empirical research needs to address questions of ‘why’ at some point in the research process. Sometimes explanations remain at the level of speculation but the research should at least point out potential avenues for further research which could explain the results, and these suggestions need to be grounded in the evidence and in the state of the art in the field. Koskinen (2010) suggests that Chesterman (2000) first adopts a Hempellian understanding of causality, according to which “[c]ausality is … a (probable) relation between particular premises and observable phenomena” (Koskinen 2010:166), and then repositions himself in a later article (Chesterman 2008a) where he supports an ‘agency theory of causation’ (Koskinen 2010:179, emphasis in original). Compared to his earlier approach, Chesterman (2008a) favours a less limited understanding of causes and follows a teleological model based on the notion of intentionally making something happen. An agency theory of causation offers a wider range of avenues for research; apart from probabilistic norms and laws, it considers Introduction  goals, intentions, motivations, and the ethics of action (Koskinen 2010:179). Koskinen suggests an alternative way of studying causality which is particularly useful for case studies and which – instead of attempting to establish correlations or causal laws – focuses on causal mechanisms, i.e. on explaining “how a particular factor can instigate a change in another factor” (2010:181, emphasis in original). She describes this approach as a “a more down-to-earth attempt to identify a plausible account of the sequence of events, conditions or processes linking the explanans and the explanandum” (ibid.:182). In her own work, Koskinen adopts a nexus model, which “is based on placing the object of study … at the centre of our attention and then trying to establish the kinds of relations it enters into and how these relations interact with it and each other” (ibid.:180). Chesterman (private communication) believes this model should be incorporated into the typology described in Chesterman (2000). The nexus model is particularly suited to case studies, a research methodology that allows us to focus on causal mechanisms rather than causal effects (see Chapter 6). We revisit the difference between mechanisms and effects in Chapter 2, Section 2.10.3. Marco (2009) proposes four (non-exhaustive) models of research in TS: (1) textual-descriptivist, (2) cognitive, (3) culturalist and (4) sociological. Marco’s classification also overlaps to some extent with the one proposed in this book; our product-oriented methods are also text-oriented, our process-oriented methods have a strong focus on cognitive processes, participant-oriented research tends to be sociological in nature and there is also overlap between cultural and contextual research. The key difference between Marco’s models and ours is that his establish a closer link between research methods and theoretical approaches or schools of thought. We prefer to encourage a looser connection between methods and schools of thought so as to offer flexibility in terms of what researchers take and discard from each methodology and from each school, and encourage creativity in terms of combining methods and theories. While the book discusses both methods and methodologies, we decided to highlight the latter in the title as the more encompassing term and because most of the chapters discuss methodologies rather than methods (see Chapter 2, Section 2.3 for a definition of these terms). Every piece of research begins with theoretical assumptions, for example, about what science is and how knowledge is constructed. Our choice of methodology depends on those assumptions as well as on our research question and/or hypothesis. The success of our methodology in addressing the research question(s) depends on how well the methods suit the research question(s) and the aim of the research. These are questions of validity and reliability which are at the basis of empirical research. Understanding such core principles is essential for any researcher no matter what methodology they adopt, and these are therefore discussed before we actually delve into methodologies per se, in Chapter 2. This chapter lays the foundations for high quality research independently of the research methodology adopted and the aspect of translation we focus on. It also discusses general ethical issues, which are then followed up in other chapters as appropriate according to the specific characteristics of the methodology.  Gabriela Saldanha and Sharon O’Brien Chapter 3 discusses how critical discourse analysis and corpus linguistics can be used to examine translated texts (including transcripts of interpretedmediated events). As explained in that chapter, critical discourse analysis is not actually a methodology but a school of thought that follows a series of principles in its understanding of language and its approach to language research. Here, we focus on the text-oriented methodology developed by Fairclough (2003, 2010) and how it has been adopted and applied in translation studies. While some linguists would argue that corpus linguistics is a research paradigm in its own right (Laviosa 2002), within the context of the present book it is presented as a methodology that can be used to pursue a wide range of research aims. Corpus linguistics and critical discourse analysis are presented as both alternative and complementary methodologies; in other words, they can be used on their own or combined. Therefore, while we take care to clearly distinguish the principles underlying each methodology, their strengths and weaknesses, and to discuss their distinctive tools and procedures, much of the discussion in that chapter concerning general principles of linguistic research applies to both methodologies. No book on research methodologies in translation studies would be complete without considering the complex nature of research involving translation quality. Such research tends to be primarily product-oriented (though alternative approaches are, of course, possible), and thus Chapter 3 also includes a discussion of research involving translation quality assessment. Chapter 4 introduces process-oriented research. We outline what the main objects of inquiry have been to date in translation process research and discuss general issues of research design, data elicitation and analysis, before focusing specifically on four methods: verbal reports, keystroke logging, screen recording and eye tracking. Chapter 5 focuses on the ‘participants’ (also called ‘agents’) involved in the process of translations, such as translators, trainers, students, commissioners and agents. This chapter discusses both quantitative and qualitative approaches to participant-oriented research and is divided into two main parts: the first discusses questionnaires and the analysis of quantitative data derived from them, and the second discusses interviews and focus groups and the analysis of qualitative data. The focus in Chapter 6 is on external – political, economic, social and ideological – factors affecting individual translators, the circumstances in which translations take place and how translations impact the receiving culture. The object of enquiry is much broader than in previous chapters and a wide range of methodologies could be used in the investigation of the very different contextual factors that can potentially be accounted for. We have chosen to focus on the case study for two reasons: first, because of its flexibility in terms of drawing from a wide range of sources of data, and second, because the label ‘case study’ is often used in translation studies research without consideration of the particular characteristics and requirements of case study as a methodology (Susam-Sarajeva 2009). While each of the chapters focuses on different research objects and methodologies, we have attempted – as far as possible – to adopt a similar structure Introduction  in each of them so as to cover consistently what we see as key aspects of any methodology: research design, data collection and/or elicitation, and analysis. In empirical social research a distinction is made between “elicitation and evaluation methods: between ways of collecting data and procedures that have been developed for the analysis of observed data” (Titscher et al. 2000:6). All chapters make a distinction between these two stages, but it is important to note that in some cases these are not two necessarily subsequent stages in a linear progression. This is particularly the case when doing qualitative research that follows an iterative process as opposed to a linear one. While many researchers use elicitation and collection as two interchangeable terms, we distinguish between the two where appropriate. Collection suggests the recording of data that already exist whereas elicitation evokes a more active generation of data which are then collected or recorded. Elicitation also suggests that a stimulus is involved, such as a text that needs to be translated according to specific instructions, and it is therefore particularly appropriate for the discussion in Chapter 4 on process-oriented methods. In our final chapter (Chapter 7) we deal with what is also generally the final stage of the research process: producing a research report. Research can be reported in a variety of formats, from conference presentations to PhD theses. Here we focus on written reports. Since many of the issues around reporting research span all our chapters and all methodologies, we discuss them at both a general as well as specific levels in this chapter. Chapter 2. Principles and ethics in research 2.1 Introduction The aim of this chapter is to highlight issues that should be of concern to all researchers and to place these in the context of translation studies research by offering examples from that domain. We commence with a discussion on ontology and epistemology, terminology and types of research. This discussion is necessarily brief: our aim is to introduce the core terms and concepts here, and we recommend that researchers turn to general works on research methodologies for fuller discussions of the issues that arise. We then turn our attention to research questions, hypotheses and types of data before considering different methodological approaches (quantitative, qualitative and mixed). The last section focuses on questions pertaining to research quality and ethics. 2.2 Ontology and epistemology There are many books on research methodologies in the humanities and social sciences which cover important philosophical questions such as How do we know what we know? or What is the truth? Here we will summarize the main philosophical questions, present the most important concepts and terms, and explain their importance for research in translation studies. It is far too easy to delve into a research project without first questioning one’s own view of the world, and, especially, of knowledge acquisition and ‘truth’. Having an appreciation for different ways of seeing the world will not only help with the decision regarding what research approach to take, but will also help us as researchers to question our own underlying assumptions, thereby hopefully strengthening our research. One of the core terms that should be understood prior to engaging in research is ontology. In social research, one way of defining ontology is as “the way the social world is seen to be and what can be assumed about the nature and reality of the social phenomena that make up the social world” (Matthews and Ross 2010:23). A key related term is epistemology, which is “the theory of knowledge and how we know things” (ibid.). Here, we follow Matthews and Ross in distinguishing, in very broad terms, three different ways of seeing the social world – objectivism, constructivism and realism – and three epistemological positions linked to these ontological categories: positivism, interpretivism and realism. These categories are somewhat convenient simplifications; in fact, there are many more than three ontological and epistemological positions, and there are also several versions of each of the positions we present here. However, analyzing these three approaches should be enough to give us an idea of the range of perspectives that can be adopted and their implications. Further reading on Principles and ethics in research 11 these questions is therefore recommended. Guba and Lincoln (2005:193) provide a helpful table that identifies five different paradigms (positivism, postpositivism, critical theory, constructivism, and participatory/cooperative) and basic beliefs associated with them concerning, for example, ontology, epistemology, methodology, ethics, inquirer posture and quality criteria. This may be a good starting point for considering the position researchers think they might be most comfortable with as this will probably influence the approach taken in research. According to Matthews and Ross (2010:24-25), objectivism “asserts that the social phenomena that make up our social world have an existence of their own […], apart from and independent of the social actors (humans) who are involved”. This position derives from the approach adopted by natural scientists when they investigate phenomena in nature and assume that the researchers’ relationship to the phenomena they study is one of objective observation. Constructivism, on the other hand, asserts that social phenomena “are only real in the sense that they are constructed ideas which are continually being reviewed by those involved in them [the social actors]” (ibid.:25). In other words, the meanings of any social phenomenon are not inherent but are ascribed to it by social actors (ibid.:28). Realism presents an intermediate position between objectivism and constructivism: it accepts that social phenomena can have a reality that is separate from the social actors involved in it but also recognizes that there is another dimension that relates to what we know about the social world as social beings. This dimension includes “structures and mechanisms that trigger or affect the social reality that can be observed” (ibid.:26). As mentioned above, each of the ontological positions described is linked to an epistemological position, that is, it entails some beliefs as to what counts as knowledge and how knowledge can be obtained. The ontological position of objectivism assumes a positivist epistemology, which asserts that social phenomena can be objectively researched, data about the social world can be collected and measured, and the resulting observations must remain independent of the researchers’ subjective understandings; that is to say, the researcher remains independent and has no impact on the data. Positivism is often linked with quantitative approaches to research and to empiricism, i.e. the collection of observable evidence (see Chapter 1). However, in postpositivist research, empiricism and objectivism are treated as distinct positions; just because research is ‘empirical’ in nature does not mean that it is ‘objective’ (Tymoczko 2007:146). In postpositivism it is held that observation and measurement are fallible, and the participation and influence of the researcher are acknowledged. As Crisafulli (2002:33) puts it, empirical facts do not exist independently of the scholar’s viewpoint; indeed, it is the scholar who creates the empirical facts of the analysis by making observable (raw) data relevant to his/her perspective. Interpretivism is linked to the ontological position of constructivism; it prioritizes people’s subjective understandings and interpretations of social phenomena and 12 Gabriela Saldanha and Sharon O’Brien is often linked with qualitative approaches to research, where the researchers attempt to explore the social world from the point of view of the actors and reflect on their own subjective interpretations. Realism is both an ontological and epistemological position. As an epistemological approach it claims that certain social phenomena exist outside the human mind and can be objectively investigated using approaches similar to those in the natural sciences. In this respect, realism agrees with positivism. However, it also recognizes the existence of invisible but powerful structures and mechanisms that cannot be directly observable but whose effects are apparent, and these effects can provide evidence of the underlying structures and mechanisms (Matthews and Ross 2010:29); Realist approaches to research might typically adopt both quantitative and qualitative tools and methods. We will not prescribe a specific ontological or epistemological framework here. In fact, the approaches outlined are not necessarily mutually exclusive, and we consider the way in which one standpoint or the other has divided researchers in the past to be unhelpful. As Guba and Lincoln state, “there is no single ‘truth’ … all truths are but partial truths” (2005:212). However, as researchers bring a number of beliefs, prior knowledge and experience to research, it is helpful to reflect on these prior to commencing research. 2.3 Research terminology Along with the terms used above, many other terms are used in research in a way that assumes general agreement about the meaning assigned to those terms. However, even seasoned researchers can use research terminology inconsistently, and this can lead to much confusion and frustration on the part of the reader and, especially, the novice researcher. Terms such as model, framework, theory, typology, concept, method and methodology often go unexplained or are used synonymously, resulting in a lack of comprehension. Here we provide some definitions for common terms, drawing mainly on Silverman (2006:13), with the exception of the definitions for ‘framework’ and ‘typology’, which are taken from Matthews and Ross (2010:34 & 112 respectively). Not everyone will agree with these definitions, which represent but one way of defining the concepts. What is important for each researcher is to carefully consider their use of research terminology, to justify the definitions used for their given purpose and to use terms consistently, while being aware and drawing attention to the fact that others might use the terms in a different way. A model is a representation of the ‘reality’ of your research topic or domain. In Chapter 1 we compared the model of translation studies research suggested by this book with the models proposed by Chesterman (2000) and Marco (2009). Note, however, that it is frequently the case that models are not made explicit in research projects and that sometimes there can be a disconnect between the assumed model and the object of investigation (Tymoczko 2007). A framework is the set of ideas and approaches that can be used to view and gather knowledge about a particular domain. As described in Chapter 3, Halliday’s systemic Principles and ethics in research 13 functional grammar is often used as an analytical framework in corpus-based translation and critical discourse analysis research. A concept is an idea deriving from a model or a framework. A theory organizes sets of concepts to define and explain some phenomenon or, in Chesterman’s words, a theory is “an instrument of understanding” (2007:1). A typology is a typical model of the way items tend to be found in relation to each other. For example, one might try to construct a typology of translation strategies used in specific circumstances. A methodology is a general approach to studying a phenomenon whereas a method is a specific research technique. In Sealy’s words, “methodology is the science of method” (2010:61). Saukko differentiates between the two concepts in the following way (2003:8; our emphasis): whereas methods refer to practical ‘tools’ to make sense of empirical reality, methodology refers to the wider package of both tools and a philosophical and political commitment that come with a particular research approach. The relation between a theory and a method is expressed by Chesterman in the following way: “methods methods are the ways in which one actually uses, develops, applies and tests a theory in order to reach the understanding it offers” (2007:1). Methods and tools are also frequently confused. Perhaps the best way to demonstrate how these terms might be applied to a domain within translation studies would be to take the example of translation process research. In this example, our model might be a particular model of cognitive processing, i.e. a representation of the phenomenon whereby the brain perceives signals, processes them and converts them into meaning and instructions. Our framework might be cognitive load, i.e. a set of ideas about brain processing capacity during a specific task, where we expect that there is a limit to the amount of information (signals) that can be processed by the human brain at any moment in time. Concepts within that framework might include the translation process, i.e. an activity the brain engages in when a human is translating from one language to another, short-term memory, long-term memory, limitations on the capacity of the brain, to name just a few. Our theory might be the MiniMax theory (otherwise known as the ‘principle of least effort’ (Krings 1986a, Séguinot 1989, Lörscher 1991), which posits that humans (translators in this case) opt for a strategy whereby they employ the highest level of cognitive processing possible and do not proceed to a deeper level of processing, which entails a greater cognitive load, unless the first level proves to be unsuccessful or unsatisfactory. Our methodology for studying this might be empirical (we will gather evidence from translators at work) and might combine both qualitative and quantitative methods such as think-aloud protocol and keystroke logging. We may wish to propose a typology of the translation strategies used to achieve the principle of least effort. Finally, the tools we might use are a voice recording device, screen recording software and a keystroke logging tool. We summarize this example in Figure 2.1. 14 Gabriela Saldanha and Sharon O’Brien Model Frarne\Nork Theory Methodology • X Model of cognitive processing • Cognitive load analysis • Concepts: translation process, STM, LTM , restrained capacity ... • MiniMax • Empirical/Experimental • Qua ntitative Methods - Eye Tracking - Keyboard Logging • Qualitative - TAP - Que s tionnaire Tools • Eye Tr acker (Hardvvare & Softvvare ) • Keyboard L ogging Softvvare • Dictaphone • On-line Questionnaire Figure 2.1 Example of Research Terminology Applied to TS Laying out one’s research domain in these terms is not always an easy task, and we expect that there will be some who do not agree with our categorization in Figure 2.1, but the exercise is worthwhile because it forces the researcher not only to think about the research terminology but also about the concepts one is subscribing to, how to communicate these to the wider research community, and, ultimately, what the researcher’s view of the world is. 2.4 Types of research There are many questions to be answered before conducting research, such as what is the research question, which method or methods are most appropriate, what kind of data will be collected, how will the data be analysed and so on. We have argued that it is worthwhile thinking about one’s epistemological framework before diving into such details. Likewise, we argue that it is important to consider what type of research we are engaging in. An initial question pertaining to type of research is what logical system it subscribes to, i.e. whether it is being conducted from an inductive or a deductive positioning. Induction involves the development of theories and hypotheses from the data collected (it moves Principles and ethics in research 15 from particular instances to general statements), whereas deduction involves the testing of existing theories or hypotheses through data (it moves from general statements to specific instances). A third position, abduction, is also possible. This position was first mentioned by C.S. Pierce in 1878; it proposes to isolate the most convincing reasons (hypotheses) from a research result and to research these hypotheses further. Johnson and Onwuegbuzie (2004:17) helpfully characterize the three as discovery of patterns (induction), testing of hypotheses (deduction) and seeking understanding by uncovering and relying on “the best of a set of explanations for understanding one’s results”. In addition to the question of logical positioning, there is the question of the nature of the research. As explained in Chapter 1, this book focuses on empirical research. Williams and Chesterman (2002:58) explain that empirical research “seeks new data, new information derived from the observation of data and from experimental work; it seeks evidence which supports or disconfirms hypotheses, or generates new ones”. This type of research is generally seen in opposition to conceptual research, which “aims to define and clarify concepts, to interpret or reinterpret new ideas, to relate concepts into larger systems, to introduce new concepts or metaphors or frameworks that allow a better understanding of the object of research” (ibid.). However, as discussed in Chapter 1, the distinction is not always clear cut and these two types of research are not mutually exclusive (see the discussion of argument-centred research designs in Chapter 3). Empirical researchers can engage in either basic or applied research. Although the distinction between these two types is not clearcut either, basic research is generally understood to mean fundamental research, the primary aim of which is to acquire new knowledge. Applied research is generally understood to mean research on practical problems, research that has an application in life. Research may also be characterized as experimental, in which case the researcher seeks to establish cause and effect relations (if X happens, then what is the effect on Y?). Such research might be carried out in a controlled environment, although this is not always practical in humanities and social science research, and is often comparative; it compares two groups and their properties of behaviour when certain variables are manipulated. It may be designed in such a way that there is an ‘experimental group’ (also known as a ‘treatment group’) and a ‘control group’. Members of the former group are exposed to some sort of ‘treatment’, or manipulation, while the latter are not. Note that the groups are not necessarily populated by humans, but can also be composed of texts, for example. The creation of control groups in translation studies research is not without challenges, however. Comparable groups of translators or translations may simply not exist. To compensate for this, Tymoczko (2002:21) suggests the use of other translations of the translated text that is under investigation or even other passages from the translated text that ‘are neutral with respect to the issues being investigated’, or the use of a corpus of parallel texts (see the discussion on obtaining comparable textual data in Chapter 3, Section 3.3.2). Experimental research in translation studies is discussed in more detail in Chapter 4. 16 Gabriela Saldanha and Sharon O’Brien Basic or applied research does not necessarily have to be experimental, though, and might also be explorative. An example of explorative research is phenomenology, where rather than asking what is the effect on Y if X happens, or what X is, the lived experience or appearance of a particular phenomenon is explored. Phenomenology is an interpretive, subjective approach to research, which is interested in gaining insights from personal experiences. For further discussion on phenomenology see, for example, O’Leary (2010) and Lewis and Staehler (2010). Research can also be evaluative, attempting to establish the value of a particular initiative once it has been implemented (summative evaluation) and the intended or unintended effects of the initiative, or it might evaluate the delivery of an initiative (formative or process evaluation). The goal of research can extend beyond that of evaluation or looking for relationships between X and Y; it can also lead to change, and this is where the term action research is applied. Action research tackles “real-world problems in participatory and collaborative ways in order to produce action and knowledge in an integrated fashion through a cyclical process” (O’Leary 2010:146). Action research is collaborative: it seeks to empower the stakeholders and moves away from the concepts of the ‘researcher’ and the ‘researched’. See Chapter 5, Section 5.7.1, for an example of action research in interpreting. Research might also be ethnographic, when it explores cultural groups “in a bid to understand, describe, and interpret a way of life from the point of view of its participants” (ibid.:116). One example is Koskinen’s (2008) study of the Finnish translation unit at the European Commission. The ethnographic approach is discussed in detail in Chapter 6. For a more detailed discussion of ethnographic research methods, see, for example, Madden (2010). 2.5 Research questions and hypotheses Before we as researchers select methodologies, we must first identify at least a tentative research question, and possibly several sub-questions, which are often refined as the research develops. The sub-questions allow the researcher to unpack the main research question into more specific, highly focused questions. Williams and Chesterman (2002) have identified many research domains in translation studies. The research question may very well ‘belong’ to one of these research domains but it may also straddle more than one domain, or explore new domains. As translation studies expands its horizons, we can expect research questions to touch on many more and varied topics than it has done to date. There are different types of research questions (Matthews and Ross 2010:57). A question might be explorative, in which case it seeks to find out what knowledge exists about a particular phenomenon. If we return to our previous example of translation processes and the theory of a MiniMax strategy, an explorative research question might be What evidence is there to show that the MiniMax strategy is employed by translators? A descriptive research question seeks to elicit data through which a phenomenon can be described in detail, e.g. What Principles and ethics in research 17 micro-strategies do translators employ when they apply the MiniMax macrostrategy? An explanatory research question is a ‘why’ question. In our example, this might be formulated as Why do translators employ the MiniMax strategy while translating? The fourth type of question is an evaluative question which seeks to understand the value of a phenomenon, e.g. What is the impact on translation quality when translators employ the MiniMax strategy? We stated in Chapter 1 that we do not believe in a clear-cut distinction between descriptive and explanatory research, and it is important to stress here again that research questions do not always fit neatly into one of the four categories above. A researcher may have an explorative question, which is then followed by a descriptive or an evaluative sub-question, for example. Questions might also have a hierarchy of sorts, with one being a primary research question, followed by one or several secondary research questions. Indeed, some primary research questions cannot be ‘operationalized’ until they are broken down into more specific sub-questions (Sunderland 2009); see below for a discussion of operationalization. Researchers will select questions based on their interest in the topic, but the question should also be one that is of interest to the community at large. Unfortunately, questions worthy of future research are not always made explicit in research publications, but it is still possible to extract questions by identifying what has not been said by authors. This requires a critical reading of research publications, where the reader considers what questions might arise from the argument being put forward and whether or not they are addressed by the author(s). It is generally accepted that research questions evolve over time. This is a normal development in the research cycle: as we become more familiar with the domain we are better able to critique our own research question and to refine it; to do so is to be recommended. This refinement frequently involves reducing the scope of the research question, or making it more specific, or introducing some sub-questions which will allow us to investigate the general question in more detail. Some research methods almost demand that questions evolve over a period of time, by the very nature of the method itself. For example, ethnographic research in general or case studies in particular might demand multiple cycles where research questions evolve as the research takes shape (see Chapter 6). Also, the use of abduction, as mentioned above, can contribute to the evolution of research questions and hypotheses. It might seem overly simplistic to say that a research question should be formulated as a question, but not doing so is a common mistake, especially among novice researchers. When novice researchers are asked about their research questions they often describe their topic instead, confusing the two and failing to formulate a question (Sunderland 2009). Williams and Chesterman (2002:57) reinforce the importance of the question type by recommending the selection of the research model based on the type of question asked. Olohan and Baker (2009:152) make an important point regarding the wording of research questions for doctoral studies: “Almost every word used in a research question sets 18 Gabriela Saldanha and Sharon O’Brien up specific expectations, some of which a student may not be able or willing to fulfill”, but this issue might also be relevant beyond Master’s or doctoral research. Sunderland (2009) echoes this point and adds that the researcher needs to understand, and explain, exactly what each word in the question means. According to Matthews and Ross, hypotheses are specific types of research questions that are not phrased as questions but as statements about relationships; they define a hypothesis as “[a] testable assertion about a relationship or relationships between two or more concepts” (2010:58, emphasis in original). A research question, then, can sometimes be rephrased as a hypothesis. If we take the descriptive research question mentioned above (What micro-strategies do translators employ when they apply the MiniMax macro-strategy?), we might express the following hypothesis in relation to this question: When translators employ the MiniMax strategy, they make use of micro-strategies that are different from those they use when they are not employing the MiniMax strategy (but see comments about the null hypothesis below). In other words, the researcher is asserting that there is a relationship between the use of the MiniMax strategy and the type of micro-strategies employed. Note that the hypothesis is not just an expression of the research question in the form of a statement. We have had to refine it somewhat in order to express it as a hypothesis, and it probably still needs further refinement. It can be illuminating to ask oneself what one’s hypotheses are, once the research question(s) has/have been formulated. In doing so, we are asking what we expect to find and the research project should aim to find evidence which either supports our hypotheses or contradicts them. Note that even if our hypothesis is not supported (or fully supported), this is still a valuable research outcome. Not all research questions can be expressed in terms of a hypothesis. In fact, those who disagree with the positivist approach to research would claim that hypotheses are reductionist devices which constrain social research (O’Leary 2010:55). Olohan and Baker (2009), in their discussion of research training for doctoral students in translation studies, comment that they favour open research questions over hypotheses for several reasons, including, for example, that an open research question provides broader scope for interrogating data from several perspectives; they encourage students to keep an open mind about the data and potential findings. Hypotheses are commonly (though not exclusively) used in situations where data can be gathered to measure each concept and where statistical tests can be executed to establish if there is a relationship between concepts. The relationships expressed in a hypothesis can be causal or associative (Matthews and Ross 2010:59). Causal relationships assert that a change in one concept (X) causes a change in the other (Y). Associative relationships recognize the influence of one concept on another. In the latter case, there might be a third factor, Z, an intervening variable (cf. Silverman 2006:289), which influences Y. Chesterman (2007) recognizes four types of hypotheses: descriptive, explanatory, predictive and interpretive. The first three can be grouped together Principles and ethics in research 19 as empirical hypotheses, whereas the interpretive kind has a different status. According to Chesterman, a descriptive hypothesis is formulated along the lines of ‘All Xs have features F, or belong to class Y’; an explanatory hypothesis states that ‘X is caused or made possible by Y’; and a predictive hypothesis is formulated as ‘In conditions ABC, X will (tend to) occur’. The interpretive hypothesis asks whether something (X) can be usefully interpreted as Y, or better understood if we ‘see it’ as something else. Chesterman (2001a) notes that classifications and categories (e.g. types of equivalence) are interpretive hypotheses in themselves. Interpretive hypotheses pertain to conceptual (as opposed to empirical) research and are “conjectures about what something means” (Chesterman 2008b:49); they are “what we use when we try to understand meaningful yet obscure phenomena” (Chesterman 2008b:56). An example of an interpretive hypothesis would be ‘translation is best conceptualized as a type of artistic performance rather than as a reproduction’. As Chesterman notes (2000:23), interpretive hypotheses are rarely presented explicitly as such, to be tested like any other hypotheses. In research that adopts a mainly quantitative approach, it is traditionally assumed that no relationship exists between two variables, and statistical tests are based on this assumption (Rasinger 2008:176). The hypothesis mentioned above would therefore be phrased as: When translators employ the MiniMax strategy, they make use of the same micro-strategies they use when they are not employing the MiniMax strategy. In other words, there is no relationship between the MiniMax strategy and the type of micro-strategies used in translation. This is called the null hypothesis and is given the notation Ho. We are usually interested in disproving the null hypothesis, in demonstrating that its opposite, or the alternative hypothesis (H1) is true. We discuss the falsification of hypotheses in more detail in Section 2.10.4. 2.6 The literature review It was mentioned above that one way of identifying interesting research questions is by performing a thorough literature review. The literature review gives researchers an opportunity to explain their motivation and potential contribution. According to Fink (2005:3), the literature review is “a systematic, explicit, and reproducible method for identifying, evaluating, and synthesizing the existing body of completed and recorded work produced by researchers, scholars, and practitioners”. Let us examine each of these qualifiers in turn: Systematic means that it is not random, i.e. that all key sources of published research on the topic have been identified, read and evaluated. Explicit implies that there is clarity regarding what works, authors, time period, domain, languages, regions, etc. have been included and, of equal importance, what has been excluded and why. Reproducible demands that everything is documented clearly, with appropriate referencing, so that any other researcher could track down the sources used and confirm the summary of the included works. 20 Gabriela Saldanha and Sharon O’Brien The literature review identifies all relevant work and synthesizes core concepts and findings. Care is needed in the synthesizing task as this does not simply involve repeating verbatim what other researchers have said, but rather summarizing the main themes, ideas, questions/hypotheses and conclusions. There are two significant challenges when synthesizing: avoiding plagiarism and deciding how to structure one’s work. Plagiarism will be discussed below in Section 2.11.8. A number of questions could be asked when considering how to structure the literature review: do you work author by author, era by era, language by language, etc., or do you amalgamate into common themes and topics? Opinions will vary on this, but generally speaking a literature review that is structured along thematic lines might be more effective and accessible than one structured chronologically and/or according to author. Arguably one of the most important features of a literature review is that it evaluates critically. For a novice researcher, who is perhaps new to doing research and to the topic itself, this is one of the most challenging aspects of the literature review. Assuming a position of modesty, the novice researcher might think that they are not in a position to criticize an author who has published one or several papers or books. However, it is the job of the researcher to critically explore previous research. We should aim to identify both strengths and weaknesses in earlier work, concepts that have not been fully investigated or researched at all, concepts that have been researched particularly well or that have been overresearched, weaknesses in assumptions, methods, research questions, and so on. There is an important comparative aspect to this commentary too: we should aim to highlight contradictory findings as well as findings which support those of previous research, and we should aim to identify differences in assumptions, theories and definitions and how these can lead to different conclusions. The literature review is an important vehicle through which researchers can identify and describe the most relevant theoretical framework(s) for their own research. Tymoczko (2007) directs attention to the interrelationship between data and theory, emphasizing that in postpositivist approaches to research the recognition of the interdependence between data and theory is essential. In the analysis of data, researchers have an opportunity to explore this interrelationship and to make explicit links to the theoretical framework(s) they have identified as being important. 2.7 Data To find answers to research questions, we need to collect appropriate data for analysis. Data can be spoken or written, non-verbal, structured in different ways, produced by individuals or groups, be factual or representing opinions, and it can include the researcher’s own reflections (Matthews and Ross 2010:181). Methods for data collection and analysis will be discussed in more detail in relation to the different methodologies presented in Chapters 3 to 6. For the mo- Principles and ethics in research 21 ment, we need to differentiate between primary and secondary data. Primary data are collected by the researcher him or herself while the term secondary data refers to collections of data, e.g. interview transcriptions, questionnaire responses, translations etc., that have been collected by other researchers and made available to the research community for analysis. Corpora could be considered in this category, so an example of secondary data for translation research would be the Translational English Corpus, a computerised collection of contemporary translational English text held at the Centre for Translation and Intercultural Studies at the University of Manchester.1 A researcher interested in analyzing some aspect of translated English could use this resource as secondary data while also creating their own corpus. Primary and secondary data might be structured in different ways. When comparing primary and secondary data, it is important to take into account that the circumstances under which data were collected, and the number and nature of the people who generated the data and the time of data collection or elicitation might vary and this may affect the comparability of the two data sets. The type of data collected is also important because it will determine whether we use qualitative and/or quantitative approaches in our research. Quantitative approaches will generate structured data which can be represented numerically and analyzed statistically, whereas the qualitative approach will generate semior unstructured data. In questionnaire surveys, for example, structured data are generated by asking the same questions to all research participants and limiting the ways in which they can provide answers (through tick boxes in questionnaires, for example). Qualitative interviews, on the other hand, generally result in semi- or unstructured data because the questions asked vary to some degree, the respondents are given some or a lot of freedom when answering and not all questions are necessarily answered. See also the discussion of quantitative and qualitative approaches to textual analysis in Chapter 3. For data that can be quantified, it is also important to take into account what kind of measurements we can apply. Rasinger (2008:25-26) distinguishes four different levels of measurement. The first level pertains to categorical scale data (also termed nominal data), where data can fall into only one category, such as ‘pregnant’/’not pregnant’. The second level is ordinal scale data, where a concept can be ranked, but where it is not possible to measure differences between each label. The example given by Rasinger here is for ranking of university lectures on a dullness scale, where it is impossible to say that ‘very dull’ is twice as dull as ‘dull’. The next level is interval scale data, where again categories can be labelled, but the difference between them is fixed. A typical example mentioned by Rasinger is the grading system used to evaluate student work. The final level is ratio scale data where, like interval data, there is a fixed value between points, but unlike interval data, ratio data have an absolute zero point. http://www.llc.manchester.ac.uk/ctis/research/english-corpus/ [Last accessed 2 December 2012]. 1 22 Gabriela Saldanha and Sharon O’Brien At what point do we have sufficient data? This is a frequently asked question which is difficult to answer because it depends on so many variables (the methodology, the research questions, hypotheses, time allocation, among others). Chapters 3 to 6 discuss data collection and address some of the relevant issues in more detail, including the length of text and corpora for analysis and the concept of saturation in participant-oriented and case-study research. In social science and humanities research in general, and translation research in particular, the trend is for researchers to work on their own. Examples of large teams of people analyzing data sets are few. Consequently, it is not always possible to analyze very large data sets. Also, the nature of the data collected in translation research – for example, written or spoken linguistic data, behavioural data, narratives – compared with the natural sciences, means that automatic analysis is challenging and not always desirable. This, in turn, tends to restrict the amount of data analyzed. While some automation is possible in translation research (see, for example, later chapters on corpus analysis and translation process research), most of the analysis is manual. Before researchers decide whether they have collected ‘enough’ data to address their research questions, they will first have to consider issues of validity and credibility, which are addressed below. It is sometimes helpful to carry out a small-scale pilot study prior to the main data collection phase. This will allow the researcher to test selected methods of analysis and will give a feeling for how much data might need to be collected to establish some level of credibility. Pilot studies are discussed in more detail, where relevant, in relation to each of the methodological approaches presented in Chapters 3 to 6. Another approach to establish whether data are sufficient is to add layers of data over time, until one sees a stabilization in the variability of results. 2.8 Qualitative, quantitative and mixed-methods approaches The approach to take to one’s research should be determined by the research question(s) and how best it/they might be addressed. The quantitative approach is associated with the positivist epistemological position we mentioned earlier while a qualitative approach is generally associated with the interpretivist position. According to O’Leary (2010:113), the qualitative tradition calls on inductive as well as deductive logic, appreciates subjectivities, accepts multiple perspectives and realities, recognizes the power of research on both participants and researchers, and does not necessarily shy away from political agendas. Each approach has specific methodologies associated with it. A qualitative apOne current example is the PACTE translation competence research group at the Universitat Autònoma de Barcelona, in Spain: http://grupsderecerca.uab.cat/pacte/es [Last accessed 2 December 2012].  Principles and ethics in research 23 proach in translation research can include critical discourse analysis, interviews, focus groups, questionnaires (see Chapters 3, 5 and 6) while the quantitative approach might be associated with corpus analysis, eye tracking, keystroke logging (see Chapters 3 and 4). It is important to point out that some methods can produce data that can be analyzed both qualitatively and quantitatively (e.g. survey data, think-aloud protocols and corpus analysis). A mixed-methods approach is the term used when several methods are used to collect or analyze data. This is often understood to mean using both qualitative and quantitative approaches. The two types of data can be collected simultaneously. Alternatively, the researcher might opt for an initial qualitative phase followed by a quantitative phase, or vice versa. The first sequence has the advantage of allowing the researcher to explore data qualitatively and to follow this exploration up with a more focused quantitative analysis of the topic or sub-topic, while the alternative of commencing with a quantitative phase has the potential advantage of exposing some trends that can then be further probed via qualitative data. Chapter 5 discusses mixed methods in participant-oriented research in more detail and illustrates different ways of ‘quantitizing’ and ‘qualitizing’ data, that is, deriving quantitative data from qualitative data and vice-versa. Guba and Lincoln (2005:201) raise an important concern regarding the commensurability of competing paradigms (e.g. positivism and interpretivism), stating that commensurability can be an issue “when researchers want to ‘pick and mix’ among the axioms of positivist and interpretivist models, because the axioms are contradictory and mutually exclusive”. As we mentioned above, in the end, the research question will dictate what the most appropriate approach is, but it is worth taking potential contradictions into account when adopting a mixedmethods approach. As Creswell and Plano Clark (2007) point out, mixed methods research is not just a way of combining qualitative and quantitative approaches, but also “a research design with philosophical assumptions” (ibid.:5). The central premise is that “the use of quantitative and qualitative approaches in combination provides a better understanding of research problems than either approach alone” (ibid.). An argument along similar lines is made in Chapter 3 in relation to the combination of corpus linguistics and critical discourse analysis. When two methods are used to collect and analyze data on the same research question, this is called triangulation, which means cross-checking the results one set of data provides with results from another set of data. This is a practice we would generally endorse, and we point to opportunities for triangulation of results where appropriate in the chapters that follow. 2.9 Research operationalization An important question to ask about the data to be collected and analyzed pertains to the unit of data. Data can pertain to either the macro or micro level (Matthews and Ross 2010:114). Macro-level data are collected, for example, from organizations, countries, systems and social entities, while micro-level data are at the level of the individual, word, or text. In the case of translation 24 Gabriela Saldanha and Sharon O’Brien research, macro-level data might pertain to professional translator associations, country-specific laws regarding language and translation, to translation practices within organizations, or to literary polysystems, to mention just a few examples. Micro-level data might pertain to the use of specific strategies in a translated text, individual translation strategies, or the length of time taken to translate a text. Tymoczko (2002) aligns macro-level research with the cultural approach to research in translation and micro-level research with the linguistic approach. In her attempt to connect the two approaches, rather than allowing them to be seen as competing and exclusive of one another, she encourages a convergence which makes use of both macro-level and micro-level analysis, with data from one type of analysis complementing and, hopefully, confirming the other. Yet another important concept is the unit of analysis. This is not the same as the unit of data. For example, the unit of data might be at the micro-level of ‘text’ and, while a researcher might analyze text in general, it is quite likely that the unit of analysis (or measurement) will be further broken down into measurable concepts such as lexical items, sentences, clauses, phrases, collocations and so on. On the macro level, the unit of data might be ‘legislation pertaining to language and translation in country X’, but the unit of analysis in this context might be specific laws or legal clauses. The unit of analysis is linked with the important concept of operational definitions or operationalization. Strictly speaking, operationalization refers to the operations involved in measuring the dependent variable. Operationalization does not pertain only to quantitative approaches to research but is equally important for qualitative approaches, where the operational definition can be thought of as an explicit and precise definition that isolates the core components of the variable under investigation. Let us go back to the example of the theory of a MiniMax strategy we used earlier which, as a reminder, posits that translators opt for the minimum amount of cognitive effort possible before proceeding to deeper levels of processing. An important question from the outset is how can I operationalize the MiniMax theory, i.e. how can I turn this somewhat abstract theory into a measurable entity? An example of how this might be achieved comes from Lörscher (1991), who equates evidence for a MiniMax strategy with a dominance of sign-oriented translation, which he defines as a transfer of the source language form into a target language form, without recourse to the sense of the text segments involved. Sign-oriented translation could be operationalized both via the translated product (where there is evidence of form substitution without recourse to sense) and via the utterances of translators produced in think-aloud protocols either during or after the translation process (for example, where a translator might say during the course of a translation: ‘Entwicklung, yes that’s development in English…, that’ll do’). Operationalization becomes even more challenging with more abstract concepts, such as culture or the status For a discussion of the problem of identifying a unit of analysis in an emerging field like translation studies, see Tymoczko (2007:153). We discuss the unit of analysis with specific reference to translation process research in Chapter 4.  Principles and ethics in research 25 of the translator. A good example is provided by Koskinen’s operationalization of ‘culture’ in the context of an ethnographic study of institutional translation (2008:40-43). Koskinen chooses a definition of organizational culture proposed by Martin (2002, in Koskinen ibid.:41) that is useful for her purposes because it describes aspects of culture, both material and ideational, that are manifested and observable, such as stories people tell, relations among people, official policies and formal arrangement (to name just a few). This definition is then related to Scott’s (2001, ibid.) systemic view of the three pillars of institutions: regulative systems, normative systems and cultural cognitive systems. Finally, Koskinen identifies methods that will allow her to explore the manifestations of culture described in the definition and belonging to those three pillars from an ethnographic perspective. Two crucial questions ought to be asked when operationalizing a concept: (1) what influence does the researcher’s beliefs and ideology have on the proposed operationalization of the concept, and (2) whether or not the tools selected can actually measure the concept the researcher wishes to measure? Take, as one example, the concept of ‘source text difficulty’, i.e. how complex the source text is and how this might impact on the translation process and/or product. There are numerous ways in which the concept ‘source text difficulty’ might be and has been measured in translation research, such as subjectively using native speakers as evaluators, using traditional readability indices, or using rhetorical structure theory (Taboada and Mann 2006). The ability of any of these methods to measure the construct or the degree to which they can do this is open to question. Operational definitions will dictate the approach for gathering data and the type of analysis that can be performed on the data. Previously, we mentioned using secondary data. There is a possibility that the operational definition used when collecting secondary data differs from the researcher’s own operational definition and, therefore, the implications of using secondary data that were collated under a different operational definition should be considered. At the very least, differences in operationalization ought to be acknowledged. 2.9.1 Measurable variables A variable “is simply something that varies … in some way that we seek to measure” (Langdridge and Hagger-Johnson 2009:40, original emphasis). This concept is used primarily in quantitative approaches to research. The dependent variable is the core concept we are trying to assess in our research question. We expect it to change when it is exposed to varying treatment. The independent variables, on the other hand, are things that we manipulate in order to see what the effect is on our dependent variable. Let us consider, for example, the research question: What is the effect on translation quality when time pressure is increased? The null hypothesis might be expressed as: There is no change in translation quality when time pressure is increased. (In fact, as we stated earlier, we expect that the opposite – alternative – hypothesis 26 Gabriela Saldanha and Sharon O’Brien is true.) Let us assume for a moment that we have found acceptable ways of operationalizing translation quality and time pressure. Then, we might expect the results shown in Figure 2.2. ,. • • • • 10... " ...... ·, ...... Figure 2.2 A theoretical model expressing the relationship between time pressure (X axis) and translation quality (nominal scale on the Y axis) In other words, as time pressure increases we expect translation quality to decrease. Of course, we would be naïve to expect such a straight linear correlation between time pressure and quality, but this figure illustrates our alternative hypothesis and our expected relationship between the dependent variable (quality) and the independent variable (time pressure). One of the challenges is how to successfully isolate dependent and independent variables so that they are not influenced by other (‘confounding’) variables. In the example above, it would not be unreasonable to expect that the complexity of the text for translation might also have an impact on quality, especially under conditions where there is significant time pressure applied. So, text complexity might be another independent variable, as might degree of experience of the translator (can more experienced translators cope better with time pressure than less experienced ones?), or degree of specialization (hypothesis: translators specialized in a specific domain will produce higher quality texts even under time pressure than those not specialized in the domain), to name just two examples. We can design a research project to investigate the effect of multiple independent variables on one dependent variable or, indeed, even on multiple dependent variables. Dependent and independent variables are typically associated with quantitative approaches to research, but qualitative approaches do not exclude them. A qualitative approach to a research question may also be able to identify depen- Principles and ethics in research 27 dent and independent variables. When items are complex and harder to measure than other items, they are typically called ‘research constructs’ rather than variables: “Constructs are unobservable variables, and therefore variables that have to be measured indirectly” (Langdridge and Hagger-Johnson 2009:40). Complex ideas such as ‘status’, ‘acceptability’, or ‘equivalence’ may be best explored as constructs in translation research. However, see Ji (2012) for examples of how the relationships between (1) texts and source texts, (2) target texts and the target language linguistic and cultural system, and (3) translation style and the historical development of the target language system, can be mapped in terms of dependent and independent variables. In relation to the second type of relationship, Ji suggests that the independent variables can be the ideological stances of the translator, and the dependent variables the particular translation strategies developed for their work (ibid.:56). On the topic of variables in translation research, another useful distinction is proposed by Chesterman (2001a), who identifies two main variable types: profile variables, which refer to aspects of the form of a translation such as stylistic or syntactic features, and context variables, which refer to aspects of the translation context and its consequences, such as text type or skopos, to name just two. 2.10 Research quality Short of reliable methods and valid conclusions, research descends into a bedlam where the only battles that are won are by those who shout the loudest. (Silverman 2006:310) The issues discussed so far in this chapter are important to ensure that research is planned in such a way that it meets high quality standards: considering our epistemological viewpoint prior to commencing the project; identifying the model, concepts, frameworks etc. with which we are working; giving due consideration to the formulation and scope of our research question and hypotheses; carrying out a literature review according to recommended practices; thinking about the approaches we wish to take and the nature of the data we wish to collect; and identifying independent and dependent variables in advance where appropriate. However, undertaking a research project also includes undertaking to make some contribution to the knowledge that already exists about a topic, and to ensure that a contribution is made, the research should meet certain quality criteria: validity, reliability and generalizability, each of which is discussed below in some detail (Section 2.10). Since some of these criteria have been criticized as not being applicable to qualitative research and an interpretivist stance, we also devote another section to alternative ways of measuring research quality (Section 2.10.4). 2.10.1 Validity Validity is a multi-faceted, and somewhat contested, topic and hence significant space is given to its discussion here. The central question around validity, according 28 Gabriela Saldanha and Sharon O’Brien to Guba and Lincoln (2005:205) is whether one’s findings are “sufficiently authentic (isomorphic to some reality, trustworthy, related to the way others construct their social worlds) that I may trust myself in acting on their implications”. The very possibility of validation in translation studies research is contested. Tymoczko (2007:155) points out that the expansion of the concept of ‘translation’ and the subsequent inclusion of a broader range of scholars has led to disagreement over the possibility of validation. However, she insists on the importance of addressing the question of validation to strengthen research methods in this field and suggests that it is important to acknowledge the limitations around claims of validity and replicability and to face these methodological problems head on (ibid.:159).4 Here, rather than attempting to argue the possibility, or impossibility, of validation in translation research, we feel it will be more fruitful to highlight the challenges and potential solutions for achieving validity in qualitative and quantitative approaches to research so that researchers can make up their own minds as to what understanding of validity they subscribe to. At the most basic level, the validity of our results will depend on the extent to which the data we collate and analyze can contribute to answering our research question. For example, let us imagine we are interested in researching translation students’ attitudes towards the teaching methods they have been exposed to in a particular university programme. We may decide to use a questionnaire to collect data. The inclusion of a question about student attitudes towards the particular university’s infrastructure (facilities, services, buildings, etc.) will not help us answer our research question, and the validity of the data accumulated through responses to that question is therefore questionable (notwithstanding the fact that the responses might give us some interesting secondary data). The definition of validity by Guba and Lincoln quoted above, according to which validity is the degree to which results match as closely as possible the real state of the world, assumes a positivistic research perspective. A less rigid (and postpositivistic) understanding of validity “concerns the extent to which justifiable inferences can be made on the basis of the evidence gathered” (Le Grange and Beets 2005:115, drawing on Messick 1989; our emphasis). The reference to inferences implies that, especially when using qualitative approaches to research, one cannot claim absolute validity. In the past, this resulted in some tension...
Purchase answer to see full attachment
Explanation & Answer:
5 pages
User generated content is uploaded by users for the purposes of learning and should be used following Studypool's honor code & terms of service.

Explanation & Answer

View attached explanation and answer. Let me know if you have any questions.

Research Methods Outlines

1. Introduction
A. The translation is not simply the process of transferring words from one language to
another but a complex interplay of linguistic, cultural, and social factors
B. This paper will explore how we can best conduct research in translation studies to
understand the translation process better.
C. We will examine the unique strengths of each method and how combining them can
create a more comprehensive understanding of the translation process.
2. Mixed Approach
A. The translation is a complex process that involves both linguistic and cultural factors. The
ultimate goal of translation is to accurately and appropriately convey the source text's
meaning in the target language
B. Qualitative research methods can provide insights into translators' experiences,
perceptions, and attitudes, helping researchers understand the factors that influence their
work
C. In contrast, quantitative research involves collecting and analyzing numerical data to
identify patterns, trends, and relationships between variables
D. While both qualitative and quantitative methods have unique strengths and weaknesses, a
mixed approach can provide researchers with a complete picture of the translation
process.
E. Another advantage of the mixed approach is that it can help researchers address each
method's limitations

F. Indeed, a mixed approach in research can lead to the development of new research
questions and hypotheses
G. For example, suppose a researcher conducts a qualitative study exploring the experiences
of professional translators working in a specific domain, such as legal translation.
H. The researcher could then use these findings to develop quantitative research questions
and hypotheses that can be tested using data collected through a survey or other
quantitative methods
I. The data collected through the survey could then be analyzed using statistical methods to
test the hypotheses generated from the qualitative study.
J. The quantitative findings could generate new research questions and hypotheses for
further qualitative investigation.
K. In summary, combining both qualitative and quantitative research methods, a mixed
approach method can effectively conduct research in translation studies
3. Conclusion
A. The paper discusses best understanding the complex translation process using a mixed
approach method that combines qualitative and quantitative research methods.


1

Research Methods

Name:

Institution Affiliations:

Course Code:

Instructor:

Date:

2

Introduction

How do we measure the quality of a translation? What factors impact a translator's
decision-making process? How can ...


Anonymous
Super useful! Studypool never disappoints.

Studypool
4.7
Trustpilot
4.5
Sitejabber
4.4

Related Tags