1. Introduction
‘Qualitative research’, claimed Lazaraton (Reference Lazaraton2003a) in a recent review of evaluative criteria, ‘has come of age in applied linguistics’ (p. 1). In the thirteen years since her own groundbreaking assessment (Lazaraton Reference Lazaraton1995) enough has changed to warrant this claim. Qualitative research (henceforth QR) has opened dimensions of insight into the processes of language teaching and learning that were not even discernible on the horizon twenty years ago, and developments in the new millennium promise even richer understandings in the future. Although I shall not advance the case for QR in this paper – its many and valuable contributions over the years have established a rich methodological pedigree from which current research benefits – it is nevertheless important to recognise that prejudice still exists: van de Ven's note that he will ‘work with an interpretive epistemology, which is not always accepted in the Netherlands’ (Reference Van de Ven2007: 112) reflects a familiar condition. The focus here, though, will be on the achievements of QR since the turn of the century.
The present paper begins with a brief introduction to the field of QR, highlighting current trends and debates in the area. This is followed by an overview of developments since 2000 based on the analysis of all papers published in 15 journals related to the field of language teaching and a more detailed description of some exemplary contributions to the field during that period. Attention then turns to the issue of quality in qualitative research. Following a review of available resources, the paper concludes with a brief comment on the prospects for QR in language teaching. A list of online resources will be found in the appendix.
2. The field of qualitative research
2.1 What is qualitative research?
Holliday (Reference Holliday2007: 1) notes that it is traditional to begin by distinguishing QR from quantitative research, an ‘unadventurous’ if necessary approach since this is how most people see it. This is certainly true and well worth bearing in mind as this section situates QR in its broader conceptual and historical context before concluding with a working definition.
In fact, the distinction between quality and quantity is just one of many convenient but rather crude alternatives such as words/numbers, subjective/objective or specific/generalisable (for a longer list see Freebody Reference Freebody2003: 3). These distinctions derive from much deeper beliefs about the nature of research itself and the world it seeks to understand, reflected in different paradigms (see Richards Reference Richards2003: 32–40 for a discussion of relevant concepts). Although the terminology varies, there is general agreement on the nature of the three dominant paradigms. Research was for a long time dominated by the first of these, (post-)positivism, based on the assumption that we can test our hypotheses about the nature of the world through a process of carefully constructed experimentation or measurement. This came to be challenged by constructivism, which rejected the objectification of knowledge and sought instead to understand, through locally situated investigation, participants' social construction of reality. A third paradigm associated with critical theory demanded an analysis of participant perspectives based on the recognition that these reflected power asymmetries produced by social and historical forces.
The emergence of new paradigms and their challenge to the hegemonic position of post-positivism led to what became known as the ‘paradigm wars’ of the 1970s and the 1980s in which rival claims were played out in various ways. The long debate was valuable in directing attention to important conceptual and methodological issues, but now to a large extent ‘peace can be regarded as having broken out’ (Bryman Reference Bryman2006: 113; for further discussion of this, see the special issue of International Journal of Qualitative Education on paradigm proliferation in educational research (vol. 19.1, 2006), which includes a useful discussion by Donmoyer (Reference Donmoyer2006) of differing views on the concept of a paradigm shift). An important contributor to this development was an attack on the sort of simplistic contrast exemplified by the quantitative/qualitative distinction (e.g. Hammersley Reference Hammersley and Brannen1992; Bryman Reference Bryman2004). Writers in our own field have adopted a range of responses to this, such as offering a more nuanced comparison with a stronger focus on practical concerns (e.g. Holliday Reference Holliday2007) or reconfiguring traditional typologies (Brown & Rogers Reference Brown and Rogers2002, for example, use ‘qualitative’, ‘survey’ and ‘statistical’ as labels), though some characterisations are more contentious. Perry's (Reference Perry and Stewart2005) suggestion, for example, that all research can be placed on a quantitative–qualitative continuum offers a useful heuristic, but one that he applies by drawing on traditional divisions.
This dissatisfaction with oppositional stances produced in the new millennium a shift towards more pragmatic approaches to QR, focusing on practical issues rather than conceptual debates. This ‘pragmatist alternative’ accepts ‘a multiplicity of positions’ (Seale et al. Reference Seale, Gobo, Gubrium, Silverman and Seale2007b: 3) and refuses to impose a single version of what counts as QR. Instead, it places the quality of research practice centre stage, emphasising the centrality of the research question (Teddlie & Tashakkori Reference Teddlie, Tashakkori, Tashakkori and Teddlie2003) and the importance of contextual factors in decision-making (Brannen Reference Brannen and Seale2007). In our own field, Dörnyei (Reference Dörnyei2007: 29–30) offers a useful characterisation in terms of ‘purist’, ‘situationalist’ and ‘pragmatic’ perspectives on traditional debates.
This shift towards more practical and contextual research issues, while not downplaying the conceptual dimension in research (see Sealey & Carter Reference Sealey and Carter2004), has rendered more abstract debates redundant and directed attention to ways in which quantitative and qualitative approaches can be integrated. This presents its own challenges (for a revealing discussion based on interviews with researchers using mixed methods, see Bryman Reference Bryman2007) and does not in itself resolve postmodernist doubts about the stability of core constructs or remove continuing positivist prejudices (see St.Pierre & Roulston Reference Pierre and Roulston2006 for a valuable introduction to some of the relevant issues here), but it nevertheless opens up fresh avenues of exploration.
An adequate definition of QR, then, must be more than merely contrastive and must seek to capture the essential characteristics of a very broad and still contested field. The definition which seems to me to come closest to achieving this is that provided by Denzin & Lincoln (Reference Denzin, Lincoln, Denzin and Lincoln2000: 4–5):
Qualitative research is a situated activity that locates the observer in the world. It consists of a set of interpretive, material practices that make the world visible. These practices transform the world. They turn the world into a series of representations, including field notes, interviews, conversations, photographs, recordings, and memos to the self. . . . This means that qualitative researchers study things in their natural settings, attempting to make sense of or to interpret, phenomena in terms of the meanings people bring to them.
Working from this definition, I have therefore included in this review research that is:
• locally situated (it studies human participants in natural settings and conditions, eschewing artificially constructed situations);
• participant-oriented (it is sensitive to, and seeks to understand, participants' perspectives on their world);
• holistic (it is context sensitive and does not study isolated aspects independently of the situation in which they occur);
• inductive (it depends on a process of interpretation that involves immersion in the data and draws on different perspectives).
This fairly open list does not exclude an element of quantification (though it is clear that an essentially quantitative study would not meet these criteria) and neither does it insist on an exclusively insider perspective, but it establishes a sense of the boundaries that apply to QR. The need for this becomes clear when the landscape of QR is considered.
2.2 The landscape of qualitative research
In Reference Denzin, Lincoln, Denzin and Lincoln1994, Denzin & Lincoln remarked, in a positive vein, that there had never been ‘so many paradigms, strategies of inquiry, or methods of analysis to draw upon and utilize’ (p. 11), though a less optimistic note had been struck two years earlier by Miller & Crabtree (Reference Miller, Crabtree, Crabtree and Miller1992), who likened the ‘quest for a useful organizational map of qualitative methods’ to ‘the quest for the holy grail’ (p. 13). Although the rise of pragmatism has rendered the desire for a clear overview less urgent, differences in terminology and categorisation can still make the field daunting for new researchers.
Terminologically, the status of paradigms is no longer problematic, given the shift away from debates in this area and a general consensus that they represent fundamental belief systems about the nature of research. Neither is there a problem about what constitutes a method, such as observation or the interview, though the term technique is sometimes used instead. The challenge lies more in deciding what counts as a core tradition, strategy of inquiry, analytic framework, or approach, all terms that are used to label the different territories within QR. The fact that Creswell uses the first of these in his Reference Creswell1998 book on QR and the last in its second edition (Reference Creswell2007) highlights the danger of assuming fixed points in this shifting world, while lists of traditions ranging from four items (Nunan Reference Nunan1992; Perry Reference Perry2005) to twenty-seven (Tesch Reference Tesch1990) testify to the impossibility of definitive mapping. Nevertheless, Prasad (Reference Prasad2005) does provide a reasonably comprehensive sketch with excellent brief overviews.
An analysis of standard introductions to QR in our field since 2000 (Brown & Rogers Reference Brown and Rogers2002; Holliday Reference Holliday2002a, Reference Holliday2007; Richards Reference Richards2003; Perry Reference Perry2005; Dörnyei Reference Dörnyei2007) produces only case study – a potentially catch-all category – as common to all, though ethnography and conversation analysis (ethnomethodology) have solitary exceptions, and action research, grounded theory, introspective methods and phenomenology also feature in more than one source. However, while phenomenology and grounded theory feature in QR generally, they do not loom large in research in language teaching and introspective methods are as likely to be quantitative as qualitative (the same might apply to case studies and even action research). Only Richards includes life history, though there is evidence of valuable work in this tradition in our field (see section 4.3 below).
To attempt a synthesis of these different categorisations would probably be futile and from a practical perspective would certainly be unproductive, but it helps to have a sense of the approaches a writer is using. In the analysis that follows I draw on the following broad categories, without claiming that the list is in any way exhaustive or exemplary: ethnographies (including linguistic ethnographies), case studies, interactional studies (and conversation analysis), introspective methods (including diary studies), life history/narrative research (and in-depth interview studies) and action research (including exploratory practice). Anyone interested in exploring relationships between these can attempt to follow the branches of Wolcott's (Reference Wolcott, LeCompte, Milroy and Preissle1992) tree, embracing twenty different traditions.
3. Review of trends in qualitative research since 2000
This section provides a (crudely) quantitative overview of QR trends since 2000 based on an analysis of leading journals in the field of language teaching. This is followed in the next section by a more developed discussion of areas of research that have received particular attention during the period.
It is worth noting at the outset that this is not the first time such an overview has been attempted. Lazaraton (Reference Lazaraton2000) analysed all data-based articles in four core journals (TESOL Quarterly, Language Learning, The Modern Language Journal and Studies in Second Language Acquisition) over a seven-year period from 1992–1997 and found only 10% were qualitative (with a further 2% ‘partially qualitative’). Yihong, Lichun & Jun (Reference Yihong, Lichun and Jun2001) also compare China and the west in terms of trends in research methods in applied linguistics, covering the years 1978–1997. Reviewing four ‘western’ journals (Applied Linguistics, TESOL Quarterly, The Modern Language Journal and International Review of Applied Linguistics), they note ‘a shift toward the qualitative direction’ and claim that ‘from the mid-1990s, the percentage of qualitative studies has been approaching that of quantitative studies’ (Yihong et al. Reference Yihong, Lichun and Jun2001: 7). The review here covers a wider range of journals, but if the ten years following the period of their study (1998–2007) are examined for the four journals they select, this trend does not seem to have continued.
Even allowing some latitude in what counts as QR, the facts are stark: TESOL Quarterly, a journal clearly sympathetic to QR, dedicated less than a third of its space to these articles (47 out of 178), while in Applied Linguistics (31 out of 196) and The Modern Language Journal (39 out of 218) they amounted to less than a fifth of the total, and IRAL included only 10 (including two mixed methods papers) in its total of 128 papers. Not all the remaining papers were quantitative, but a representation of less than 18% of the total does not suggest a narrowing of the gap between qualitative and quantitative studies. The period covered by Yihong et al. (Reference Yihong, Lichun and Jun2001) was an important one in terms of the development of QR in language teaching and the shift they remark on was notable at the time, but the following analysis of trends in the new millennium suggests that things have settled into a situation where QR has a solid – though minority – presence in leading journals.
Any choice of ‘representative’ journals is bound to be contentious, but the overview that follows draws on 15 international publications, all with a research dimension and all of which at least have a reasonable claim to importance. In fact, the list includes all but one of the journals featuring in Egbert's (Reference Egbert2007) quality analysis, Studies in Second Language Acquisition being surveyed but excluded because of the predictable absence of qualitative studies (though the case is less clear-cut, I excluded Language Testing for the same reason). Some of the journals have a more local geographical relevance than others (though journals related to a specific country were not included), while some have a specialist focus, and they range from those with a heavy research focus to others with a more practical orientation. They also differ to some extent in terms of length of article, but the only journal excluded on this basis is Language Learning Journal because with an article length of around 4–6 pages it allows sufficient space only for reports on research and excludes an important methodological dimension. The full list is as follows: TESOL Quarterly, The Modern Language Journal, RELC Journal, JALT Journal, Prospect, Asian Journal of English Language Teaching, System, ELT Journal, Language Teaching Research, Language Learning, Journal of Second Language Writing, International Journal of Bilingual Education & Bilingualism, Applied Linguistics, English for Specific Purposes and International Review of Applied Linguistics (IRAL).
In surveying these journals I have included all papers, excluding editorials, introductions, forums, comment papers, reviews, etc. This means that when QR papers are removed what remain are not necessarily just quantitative studies, but since my aim here is to present a picture of QR representation in general rather than to compare quantitative and qualitative studies, this seems legitimate. The analysis itself makes no judgement on the quality of the contributions and is based on the broad definition of QR provided in the introduction. This seems to work straightforwardly in most cases, though the term itself cannot always be relied on. Takahashi's (Reference Takahashi2005) title, for example, promises a ‘qualitative analysis’ but this belies the experimental study that follows. Where the analysis depends entirely on the analysis of discourse, I have made decisions on the basis of whether the interest is primarily in aspects of the discourse itself (i.e. essentially linguistic), in which case I have excluded it, or on the individual or social dimension, in which case it has been included. Mixed methods approaches are included where the qualitative aspect predominates, though they do not seem to have yet made a significant impact in language teaching research despite their potential, exemplified in Basturkmen, Loewen & Ellis's (Reference Basturkmen, Loewen and Ellis2004) study of the relationship between the beliefs and classroom practices of three teachers.
If anything emerges clearly from a study of papers published in these journals between 2000 and 2007 it is that, with a couple of notable exceptions, they have remained remarkably consistent in terms of the extent to which they feature papers involving QR. In terms of representation, the journals fall very roughly into two equal groups: those where QR papers take up less than 10% of the total (International Journal of Bilingual Education & Bilingualism, IRAL, Language Learning, RELC Journal, English for Specific Purposes, ELT Journal, Journal of Second Language Writing and System), and those in which either roughly one in five of the papers (The Modern Language Journal, Applied Linguistics, Prospect and JALT Journal) or around a quarter of the papers involve QR (TESOL Quarterly, Asian Journal of English Language Teaching and Language Teaching Research). It is perhaps not surprising – and quite encouraging – that most of the journals with more ‘general’ coverage fall into the second category.
There are other encouraging signs. Both Prospect and ELT Journal, for example, show signs of increasing the proportion of QR papers over the last couple of years, and while questionnaire surveys have always featured very prominently in JALT Journal there has been a noticeable increase in QR contributions over the last few years, rising from about one in ten papers up to the end of 2004 to over one in three after that. One of the most interesting developments is to be found in the Journal of Second Language Writing, where more QR papers were published in 2006 and 2007 than in the previous six years combined, with the emergence of individual case study particularly prominent.
Case studies are widely represented and feature most prominently in the Journal of Second Language Writing, but other approaches are also particularly associated with specific journals. For example, The Modern Language Journal seems to favour discourse-based papers, which represent around two-thirds of the total QR papers in the journal. These also feature strongly in Applied Linguistics, where papers based on introspective methods are rare, in contrast to the generous provision in ELT Journal and Prospect. Intriguingly, action research and exploratory practice do not seem to be well represented generally, with a significant presence in only Language Teaching Research, where exploratory practice is allotted its own separate section. The same might be said of ethnographic projects, which feature in nearly all the journals but far from prominently. TESOL Quarterly, where a quarter of the QR papers fall into this category, is a notable exception.
The qualitative landscape seen through the lens of journal readers, then, might seem fairly bland: QR is reasonably well represented across a range of traditions, with none emerging as predominant and no evidence of a trend in any particular direction. However, a more careful examination of the sort of work being published in QR, in these journals and in other forms of output, reveals that new territory has been opened up through this form of inquiry. When compared with QR published in other fields, for example, it becomes immediately apparent that the use of teacher and student journals as a data source is a distinctive feature of research in language teaching, and the development of introspective approaches represents a notable contribution made by our field. The next section will highlight areas of research in language teaching (including language support in L1 situations where appropriate) that have benefited particularly from the strengths of QR. It is necessarily selective and based on what seems to me to be most interesting in terms of QR, and I have deliberately tried not to describe territory already mapped by other articles in this journal.
4. Qualitative research contributions to research since 2000
4.1 Approaches to teaching
Researchers have continued to build on QR's important contribution to our understanding of what happens in language classrooms, and studies of teachers' perceptions of communicative language teaching and their approaches to it continue to proliferate (e.g. Miller & Aldred Reference Miller and Aldred2000; Mangubhai et al. Reference Mangubhai, Marland, Dashwood and Son2004, Reference Mangubhai, Marland, Dashwood and Son2005; Nazari Reference Nazari2007). The contradictory results of research on Communicative Language Teaching (CLT) and teacher stereotypes are thought-provoking. For example, while Wu & Fang's (Reference Wu and Fang2002) study of teachers in China found that their perception of teaching as one-way transmission of knowledge inhibited their approach to CLT, Xinmin & Adamson's (Reference Xinmin and Adamson2003) in-depth study of a single Chinese teacher directly challenged this stereotype, exposing a dynamic interplay of experience, beliefs and practice, findings supported by research in Vietnam (Ha Reference Ha2004) and Korea (Mitchell & Lee Reference Mitchell and Lee2003). This is an area which would repay further research.
Research on student writing and feedback on this has benefited from QR's focus on processes and relationships from the perspectives of those involved (for a review of ethnographic research in L2 writing, see Ramanathan & Atkinson Reference Ramanathan and Atkinson1999). There is evidence, for example, of a developing interest in the influence of learning biography (Leibowitz Reference Leibowitz2005), beliefs and previous writing experience (Yasuda Reference Yasuda2005) on the process of composition. In-depth interviews feature prominently in data collection here, while between-method triangulation is particularly rich in feedback studies. In their study of the relationship between novice tutors and ESL tutees, for example, Weigle & Nelson (Reference Weigle and Nelson2004) used online discussions from the course, videotapes of tutoring sessions, retrospective interviews with tutors and tutees, and the tutor's final reflective papers for the course. Hyland's (Reference Hyland2000) study of autonomy in feedback provides a useful illustration of how an impressively wide range of methods (questionnaires, interviews, think-aloud protocols, classroom observation and documents) can be applied over the length of a three-month course, while Hamp-Lyons' (Reference Hamp-Lyons, Hyland and Hyland2006) presentation of drafts, lesson exchanges and interview data in her case study is exemplary. Also methodologically interesting are Asaoka & Usui's (Reference Asaoka and Usui2003) use of student journals and interviews to provide insights into the problems with writing; Zhu's (Reference Zhu2004) combination of genre analysis and focus group interviews to explore professional perspectives on the writing of business faxes by Chinese students; and Li's (Reference Li2007) exploration of the first draft of a chemistry student's paper based on the student's process logs, his developing text, his bulletin board exchanges and post-hoc interviews.
Conversation analysis has established a major presence in the area of speaking (but for a broader overview see McCarthy & O'Keeffe Reference McCarthy and O'Keeffe2004) and has been comprehensively reviewed by Seedhouse in this journal (Reference Seedhouse2005). The literature is marked by differences of emphasis and debates on the precise nature of its contribution continue (Kurhila Reference Kurhila2006; Hall Reference Hall2007a, Reference Hallb; Hua & Seedhouse Reference Hua and Seedhouse2007; Seedhouse Reference Seedhouse2007; Hellermann Reference Hellermann2008). Research has extended into different language environments (e.g. Buckwalter Reference Buckwalter2001 for Spanish; Liebscher & Dailey-O'Cain Reference Liebscher and Dailey-O'Cain2003 for German) and is opening up new aspects of interaction such as humour (Bell Reference Bell2005) and laughter (Reutzel Reference Reutzel2003).
Multiple methods have also been used to explore aspects of classroom interaction. Ohta & Nakaone (Reference Ohta and Nakaone2004), for example, use recordings of over 30 hours of classroom talk, observation and documentary evidence to open up student questioning, while Tan (Reference Tan2007) revealed problems with teachers' questioning behaviour through lesson observation, semi-structured interviews and focus group discussions. This seems to be an area with considerable scope for further exploration, but even more exciting is the recent growth of developmental studies. Hellermann (Reference Hellermann2006), for example, used longitudinal microethnography to trace a shift from peripheral to engaged participation in classroom literacy events and, in Hellermann Reference Hellermann2007, followed six successful learners over a period of 18–27 months to identify the development of conversational practices in student dyads. Along similar lines, Cekaite (Reference Cekaite2007) used microanalytic and ethnographic methods to trace a child's emergent L2 interactional competence during the first year in a Swedish immersion classroom. The findings of these papers suggest that there are significant benefits to be gained from further QR of this kind.
Although reading and listening have been less well-served by QR, the range of approaches used by research on the former suggests that it has been under-exploited. Kamhi-Stein (Reference Kamhi-Stein2003), for example, brings together think-aloud protocols, open-ended interviews, self-assessment inventories and reading comprehension measures to explore the relationship between L1 and L2 reading strategies, revealing how attitudes to home language influenced reading behaviour, while Miller, Mitchell & Brown (Reference Miller, Mitchell and Brown2005) develop a powerful picture of the literacy development of African refugees in Australia, relating this to social backgrounds. Martin's (Reference Martin2003) microethnographic study of reading practices in an up-river school in Brunei describes the way in which members of three different ethnic groups position themselves and accomplish literacy events, providing an excellent illustration of how close analysis of a single lesson can generate valuable insights.
Think-aloud methods dominate QR on listening, though treatments vary considerably. Vandergrift's (Reference Vandergrift2003) study of the listening strategies of grade seven students of French, for example, begins with quantitative analysis but goes on to provide a convincing justification for the inclusion of qualitative analysis and a rich sample of responses which provide model support for this. Goh (Reference Goh2002) does not include the same element of quantification in her study of Chinese ESL learners' listening strategies and tactics, but her categorisation of different tactics is supported by samples and comments, while Farrell & Mallard (Reference Farrell and Mallard2006) provide extracts from the classroom with briefer summaries of student comments. This concentration on think-aloud approaches is undoubtedly effective, but research on reading suggests that other approaches also have much to offer and it would be unfortunate if the neglect of these in listening research is allowed to continue.
Qualitative approaches have also opened windows onto aspects of the classroom community (e.g. Creese Reference Creese2002; Duff Reference Duff2002) and contributed to an improved understanding of pedagogic practice (e.g. Walsh Reference Walsh2002, Reference Walsh2006b; Carless Reference Carless2004; Dufficy Reference Dufficy2005; Hammond & Gibbons Reference Hammond and Gibbons2005; Lacorte Reference Lacorte2005; Gibbons Reference Gibbons2006; Richards Reference Richards2006a; Pinter Reference Pinter2007; Sakui Reference Sakui2007). Gieve & Miller (Reference Gieve and Miller2006) provide a discussion of relevant issues and Chavez's (Reference Chavez2006) study of experienced teachers of German illustrates how narrow but very rich data sets can be exploited to excellent effect.
Useful work has been done on teacher collaboration, where the combination of interviews and the analysis of classroom extracts has proved particularly effective. In a methodologically exemplary study, Creese (Reference Creese2005, Reference Creese2006) offers revealing insights into the relationship between subject and language teachers in UK schools. Researchers such as Perry & Stewart (Reference Perry and Stewart2005), Aline & Hosada (Reference Aline and Hosada2006) and Carless (Reference Carless2006) have also advanced the understanding of team teaching in countries such as Japan, South Korea and Hong Kong, as well as contributing to the development of good practice. However, a more negative picture emerges from Sato & Kleinsasser's (Reference Sato and Kleinsasser2004) study of Japanese high school EFL teachers, which found that teacher collaboration served only to reinforce existing practices, eroding teachers' motivation.
Extending beyond the classroom, an area that would repay further investigation is the relationship between home and school. Xuesong (Reference Xuesong2006) has used biographical interviews to explore Chinese parents' involvement in their children's language development, while Neville-Barton (Reference Neville-Barton2002) used questionnaires and interviews to study the impact of family lives on language learning of eight immigrant students from China, Korea and Japan. These are areas where greater understanding might make a significant contribution to maximising language learning potential, as Wallace's (Reference Wallace2005) research on bilingual learners at both home and school indicates.
4.2 Identity and socialisation
One of the major themes to emerge since 2000 is that of identity, and QR has made a significant contribution to our understanding of different aspects of the language learning experience. Major studies and collections have explored L2 and multilingual contexts, focusing on learners of different ages and backgrounds in different contexts (e.g. Norton Reference Norton2000; Toohey Reference Toohey2000; Day Reference Day2002; Miller Reference Miller2003; Pavlenko & Blackledge Reference Pavlenko and Blackledge2004; Block Reference Block2007), and there is every indication that this interest will continue to grow. Observation and interviews, sometimes combined with discourse analysis, have also been used to study the process of classroom socialisation in a variety of settings and levels including a kindergarten in the US (Hawkins Reference Hawkins2005), a bilingual school in Italy (Mickan Reference Mickan2006) and a university in Germany (Chavez Reference Chavez2007), while the influence of cultural and educational background on performance and classroom behaviour has also proved a fruitful area for qualitative researchers (e.g. Gao, Li & Li Reference Gao, Li and Li2002; Connor & Rozycki Reference Connor and Rozycki2006). Rampton's (Reference Rampton2006) exploration of the use of German both inside and outside the classroom opens new avenues of investigation and his concept of ‘crossing’ (Rampton Reference Rampton1995) has already influenced later studies (e.g. Stroud & Wee Reference Stroud and Wee2007).
QR has also opened a revealing window on the world of the immigrant or sojourner. The focus has typically been on particular groups (e.g. Gordon Reference Gordon2004; Menard-Warwick Reference Menard-Warwick2005) or educational settings (e.g. Morita Reference Morita2004; Vickers Reference Vickers2007), but Gordon's (Reference Gordon2006) study of strains on the lives and careers of teachers in Japan who work with marginalised youth throws light on an area that would repay further research. Study abroad has also emerged as an area of interest, producing work on identity construction (Pellegrino Reference Pellegrino Aveni2005), engagement with the host culture (Wilkinson Reference Wilkinson2002; DuFon Reference DuFon, DuFon and Churchill2006; Iino Reference Iino, DuFon and Churchill2006), and cultural adjustment (Bacon Reference Bacon2002; Gu & Schweisfurth Reference Gu and Schweisfurth2007), among other areas (see DuFon & Churchill Reference DuFon and Churchill2006 for examples).
Research on teacher identity has thrown up interesting perspectives, including professional identity formation (Alsup Reference Alsup2005), racialised identities (e.g. Hammond Reference Hammond2006; Motha Reference Motha2006), religious beliefs (Varghese & Johnston Reference Varghese and Johnston2007) and relationship to place (Elbaz-Luwisch Reference Elbaz-Luwisch2004). Non-native-speaker teachers have received particular attention: dimensions such as cultural knowledge (Lazaraton Reference Lazaraton2003b), ‘legitimacy’ (Golombek & Jordan Reference Golombek and Jordan2005; Tsui Reference Tsui2007) and relations with native-speaker teachers (Park Reference Park2007) have all been explored. Kurihara & Samimy's (Reference Kurihara and Samimy2007) study of the impact of an American teacher training programme on the beliefs and practices of eight Japanese teachers on their return to Japan is also worth noting (for complementary perspectives, see McKay Reference McKay2000; Lamie Reference Lamie2001).
Methodologically, one of the most encouraging features of research on identity and socialisation has been the appearance of longitudinal studies, from two or three years (e.g. Maguire & Graves Reference Maguire and Graves2001; Gordon Reference Gordon2004; Golombek & Jordan Reference Golombek and Jordan2005) up to five or six years (Caldas & Caron-Caldas Reference Caldas and Caron-Caldas2002; DuFon Reference DuFon, DuFon and Churchill2006). Ortega & Iberri-Shea's (Reference Ortega and Iberri-Shea2005) critical review of longitudinal studies published in 2002 and 2003 shows that work is being done along these lines, but there is still a pressing need for more longitudinal qualitative studies.
4.3 Narrative/Lives
Although this tradition seems to have been largely ignored in introductions to QR in language teaching, it has produced some interesting studies. In a two-year longitudinal case study of a Chinese student's language learning, Gao (Reference Gao2007) makes an eloquent case for the value of seeking to understand the learning experience from a biographical perspective (see also Menard-Warwick Reference Menard-Warwick2005), and teacher biographies have thrown light on relationships to cultural and educational contexts (e.g. Doecke Reference Doecke2004; Tsui Reference Tsui2007). The role of narrative inquiry as a tool in language teacher development has also been explored (Golombek & Johnson Reference Golombek and Johnson2004), while Cowie's (Reference Cowie2006) inclusion of an element of autoethnography as an English teacher and Japanese learner opens up fascinating possibilities.
Despite some useful discussions of theoretical and methodological issues (e.g. Bell Reference Bell2002; Pavlenko Reference Pavlenko2002, Reference Pavlenko2007), approaches to life history are very varied, as a comparison of papers by Simon-Maeda (Reference Simon-Maeda2004), Cheung (Reference Cheung2005) and Hayes (Reference Hayes2005) reveals. In fact they seem to have little in common beyond the length of interviews (roughly 2–3 hours) and the inclusion of interviewee profiles. Cheung's approach to understanding phases and changes in teacher careers is characterised by summaries (often tabular) and discussion of her findings but relatively limited direct representation of teachers' voices. However, she provides methodological details that are missing from Simon-Maeda. The latter's use of quotation is extensive and teacher voices emerge clearly, but the result is conceptually thin for a paper that claims to contribute to a ‘theoretically informed debate’ (Simon-Maeda Reference Simon-Maeda2004: 431). Only Hayes addresses fundamental methodological issues such as trustworthiness and the researcher's place, describing data collection procedures fully though not extending this to analytical decisions. His analysis is arranged thematically, with a blend of summary and quotation that captures teachers' voices more fully than Cheung but less richly than Simon-Maeda.
4.4 Other developments
This necessarily selective overview has omitted two aspects of QR which are becoming increasingly prominent. The first, teacher beliefs, has already been reviewed in this journal (Borg Reference Borg2003a) so a briefer overview here would be redundant. The growth of research using introspective methods also needs to be noted, but in the space available it would be impossible to do justice to the complex relationship between quantitative and qualitative approaches here. Diary studies are less problematic and have a history dating back to Bailey's pioneering work (Reference Bailey, Seliger and Long1983, Reference Bailey, Richards and Nunan1990). Most focus on learner strategies (e.g. Halbach Reference Halbach2000; Hart Reference Hart2002; Huang Reference Huang2005) or teacher reflection and learning (e.g. Marefat Reference Marefat2002; Towndrow Reference Towndrow2004; Lee Reference Lee2007), though they have also been used to explore aspects such as time management (Ho Reference Ho2003) and differences between teacher and learner agendas (Zhanjiang Reference Zhanjiang2006). What seems to be lacking, though, is an extended and rigorous treatment of relevant methodological issues.
An exciting development in QR in this period has been the emergence of linguistic ethnography, focusing on the relationship between language and social life. There seem to me to be at least three reasons why this should be welcomed by researchers in our field: at the heart of the tradition is an attempt to bring together linguistics and ethnography; the process of emergence brings to the surface fundamental issues relating to the nature of QR; and the quality of core work in the tradition so far is exemplary. The relationship at the core of the tradition between ‘tying ethnography down’ and ‘opening linguistics up’ is examined in a seminal discussion paper (Rampton et al. Reference Rampton, Tusting, Maybin, Barwell, Creese and Lytra2004: 4) and the debate on this is taken up in a special issue of the Journal of Sociolinguistics where, for example, Sealey (Reference Sealey2007) and Blommaert (Reference Blommaert2007) engage on issues of theory.
Theoretical and methodological issues are brought to the fore in three key studies in this tradition. The first of these, Creese (Reference Creese2005), has already been discussed, though it is worth underlining here the effectiveness of the way data from interviews, classroom exchanges and fieldnotes are presented. One of the strengths of Rampton's (Reference Rampton2006) study of teenagers in an inner-city school is its richness of detail in terms of both data presentation and analysis. The ‘methodological reflections’ in chapter 10 not only offer insights into the author's decision-making but also take the reader into Rampton's confidence in matters of methodological choice. It serves as a model for the sort of methodological transparency that is often demanded but rarely delivered. Maybin's (Reference Maybin2006) exploration of children's verbal practices inside and outside the classroom raises useful questions about analytical positioning and the extent to which researchers need to fashion individual approaches to suit their interpretive needs. Drawing on linguistic ethnography and poststructuralist theory for her analytical framework, Maybin is able to develop her own analytical style, revealingly exploratory in earlier chapters but increasingly confident and persuasive as the book develops. Methodologically as well as analytically, these three books are important resources for researchers in all qualitative traditions, making productive use of the greater space available in book-length studies when compared with research articles.
5. Quality in qualitative research
5.1 Checklists and standards
In spite of the shift away from paradigm wars and towards the more pragmatic perspective described in the first section of this paper, St.Pierre & Roulston (Reference Pierre and Roulston2006) note in their introduction to a special issue of the International Journal of Qualitative Studies in Education on the state of qualitative inquiry that the politics of educational research are such that researchers ‘are encouraged to develop a “culture of science” and that science does not seem to be qualitative’ (p. 679). In this section I consider how QR has responded to this challenge, beginning with an introduction to guidelines for conducting research, moving on to consider examples of published research which illustrate core quality issues, and finally considering lines of response to current challenges.
In matters of guidance on good research practice, language teaching is perhaps uniquely well-served in having guidelines provided by TESOL Quarterly (2003) which focus specifically on case studies, conversation analysis and critical ethnography. The American Educational Research Association's ‘Standards for reporting on empirical social science research in AERA publications’ (AERA 2006) are more wide-ranging and are helpfully structured, while Long & Godfrey (Reference Long and Godfrey2004) provide a briefer ‘evaluation tool’. As resources these guidelines are admirable, but the strength deriving from their precision is at the same time a source of potential concern. QR is a heterogeneous field resistant to categorical specification and perhaps best approached from the more general perspective of ‘common notions’ (Freeman et al. Reference Freeman, deMarrais, Preissle, Roulston and Pierre2007), a point eloquently made by Shohamy (Reference Shohamy2004) in her discussion of boundaries and structure as part of her response to the TESOL guidelines.
Holliday's (Reference Holliday2004a) response to the same guidelines reflects pragmatic developments in QR, making a fundamentally important point: the research activities featuring in the case study guidelines ‘should be ever expandable to do what researchers need it to do to answer their research questions’ (p. 732). His association of this with the need for thick description (for an illuminating brief discussion of this concept, see Ponterotto Reference Ponterotto2006) and transparency cuts to the heart of validity issues in QR, and his illustrative case brings out an often neglected dimension in the research process: the relationship between flexibility, responsiveness and transparency of representation.
Checklists or guidelines are not in themselves guarantees of quality in QR and ‘technical fixes’ in general (Barbour Reference Barbour2001) are no substitutes for systematic, sensitive and careful analysis. As Atkinson & Delamont note (Reference Atkinson and Delamont2006: 750), ‘[a] few interviews, a handful of focus groups or a period of autobiographical navel-gazing can never generate faithful renderings of the complexity of social life and its multiple forms of social action’. Hammersley (Reference Hammersley2006: 6) associates briefer time spent in the field with a move away from an older anthropological model of ethnography, a point also made by Holliday (Reference Holliday2004b) in his appeal for more creative forms of data management. We need to be sensitive to this, especially in the case of micro-ethnography, where attention to micro-analysis of action can distract attention from the bigger picture. This challenge is particularly salient in our field, where the bedrock exerts a magnetic pull towards linguistic analysis. While the TESOL guidelines rightly respond to Lazaraton's (Reference Lazaraton2003a) criticisms of an earlier failure to accommodate conversation-analytic perspectives, her fundamental point about the need for greater sensitivity to standards of judgement in evaluating the validity of qualitative accounts remains pertinent.
Issues of quality are of course wide-ranging (Borg Reference Borg2004 provides a succinct introduction to key topics), but the challenge identified by St.Pierre & Roulston requires that our evaluative antennae remain tuned to issues of rigour and validity, however conceived. The first phase of standards in QR was necessarily a general one and has produced sound guidelines; the next is likely to involve deeper engagement with theoretical and procedural issues. The remainder of this paper will indicate where current fault lines are to be found and what steps are being taken to respond to them.
5.2 Issues of quality in published work
Since terms such as validity and reliability remain contested in QR, this section will discuss quality issues in terms of Lincoln & Guba's (Reference Lincoln and Guba1985: 289–331) alternatives, credibility, transferability, dependability and confirmability, which feature in discussions of the topic in applied linguistics (see Brown Reference Brown, Davies and Elder2004: 494–495 for a useful brief characterisation). The issue of transparency will also be considered.
Dependability in QR involves an interrogation of the context and the methods used to derive data. Methodological positioning cannot be separated from the way in which data are presented and the nature of the claims made, and unsurprisingly practices vary widely. Perpignan's (Reference Perpignan2003) paper on written feedback, for example, demonstrates that it is not enough merely to collect data using a wide range of methods if details of how they are combined and analysed are not provided. By contrast, in a necessarily brief discussion of methodology as part of a much shorter paper, Tardy & Snyder (Reference Tardy and Snyder2004) relate their methodological decisions to the nature of the claims they wish to make, providing useful additional detail in brief appendices and allowing the participants' voices to emerge clearly in their analysis, thus supporting their methodological positioning. Hall's (Reference Hall2008) reflections on his diary study come towards the end of his paper and are made explicit in the section heading: ‘Did my research methodology affect the data?’. These examples of good practice are not isolated but it is probably fair to say that methodological interrogation could be more widespread.
One area in need of particular attention in this respect is that of interviewing. Interviews feature prominently in QR in language teaching, but there are as yet few signs that researchers have taken note of developments in the wider field, Pavlenko's (Reference Pavlenko2007) excellent discussion of autobiographic narratives as data and Talmy's (Reference Talmy2008) AAAL (American Association for Applied Linguistics) colloquium being notable exceptions. There is a growing literature on the importance of treating interviews as interactionally co-constructed events in which participant identity and positioning have significant analytical implications (Baker Reference Baker and Silverman1997, Reference Baker, Gubrium and Holstein2002; Nijhof Reference Nijhof1997; Rapley Reference Rapley2001; Cassell Reference Cassell2005; Roulston Reference Roulston2006; Wooffitt & Widdicombe Reference Wooffitt, Widdicombe, Drew, Raymond and Weinberg2006), while problems of memory (Gardner Reference Gardner2001) and misrepresentation (Sikes Reference Sikes2000) represent particular challenges. Yet analysis of interviews in our field still tends to treat them as reports rather than accounts, relying on unproblematised thematic analysis.
There is no shortage of illustrative examples of this tendency, but I choose Palfreyman (Reference Palfreyman2005) because it is a valuable interview-based study from a leading journal in our field. Although it explicitly refers to issues of representation and recognises the relevance of personal relationships, these aspects are not examined in terms of the interviews themselves. Neither is any indication given of how analysis was approached, even though in other respects (e.g. data presentation) the paper is exemplary.
Confirmability in QR depends on making the data available to the reader, and this in turn depends on transparency of representation. An impressionistic assessment, based on the fairly extensive reading involved in this review, is that there is a trend towards richer representation, with participants' voices and perspectives emerging clearly, though in ethnographic studies word limits may restrict the extent to which relationships can be fully represented. Of course, even shorter papers can provide excellent coverage (see, for example, the range in Borg Reference Borg2001a), while extended accounts can integrate different data sources to powerful effect (e.g. Creese Reference Creese2002). Exceptions are typically justified in terms of the aims of the research. Huang's (Reference Huang2003) ethnographic study, for example, provides no examples from fieldnotes and only one extract of classroom interaction, though its detailed descriptions of the pedagogic context and procedures, and its focus on materials and the language produced as a result, properly reflect the aim of the paper.
Transparency has been described as ‘fundamental to good research practice’ (Dale Reference Dale2006: 79), but it is perhaps the most difficult aspect for a researcher to manage and an outsider to judge, not least because of the complex issues which may underlie representational decisions. With no access to the decision-making processes of other researchers, I illustrate this with an insider's perspective on two research outputs, focusing on what would seem to be on the surface one of the most straightforward issues: researcher representation in the data.
In their paper on justifying outcomes in QR, Edge & Richards claim that authentication ‘involves making available an appropriate selection of the records of the research process’ (Reference Edge and Richards1998: 351), leaving open the issue of what counts as appropriate. In his book on identity in professional interaction, Richards (Reference Richards2006b) appears in two of the three groups featured, in one as himself and in the other under a pseudonym. His justification for the latter is that he is a member of the group whose exchanges were originally recorded by another researcher. Nevertheless, this effectively produces three voices in the book, which the reader may not easily identify: author, author-as-researcher, and group member. In a paper on computer-mediated cooperative development published in the same year, Edge (Reference Edge2006) assigns individual names to four different participants but does not identify himself as representing two of these. The justification for this would seem to be that, since the discourse roles in these exchanges are fixed, personal idiosyncrasies are not relevant to the analysis and would be distracting. Nevertheless, the decision produces a confusing picture in terms of the number of participants involved and commits the analyst to creating an artificial distance between himself and his pseudonymic representations.
It should be stressed that in neither of these cases does the use of a pseudonym influence the analysis or affect the results, but the issue of transparency remains awkwardly present and essentially unresolved. The broader issue of the relationship between the researcher's voice and those of other participants is succinctly discussed by Holliday (Reference Holliday, Miller and Thompson2002b), who is careful to reflect the complexity of the dilemmas involved rather than suggesting comfortable solutions.
Transparency is closely related to the requirement for researchers to establish the credibility of their interpretation, and though it seems excessive to insist that ‘every ethnography be accompanied by a research biography’ (Ball Reference Ball1990: 170), it is reasonable to expect evidence that alternative interpretations of the data have been considered. Constant comparison within the data set (looking for new relationships, categories, etc.), the search for negative evidence and the use of member validation are important steps in this process, so it is disappointing that reference to these procedures is rare. Ha (Reference Ha2004) offers a welcome exception, giving transcripts of interviews to participants, while Guerrero (Reference Guerrero2003) refers to member validation and demonstrates in his analysis that he is sensitive to the discrepant case (p. 661). These, though, are exceptions rather than the rule, and even book-length studies often fail to provide details of the author's engagement with methodological and analytical issues.
To meet the demands of transferability, the research needs to provide a sufficiently rich description of the project for readers to assess how far it might apply in their own context, and here the picture is a much happier one. Though few accounts are as detailed as that of Duff (Reference Duff2002), the level of local contextualisation is generally high. Van Lier (Reference van Lier, Bailey and Nunan1996) offers a particularly engaging and revealing account of a research project in Peru, shot through with a tension between the demands of research and a commitment to pedagogic action. The quality of the author's descriptions of the context (derived from fieldnotes), the feeling for place and the honesty of his reflections convey a sense of constant interrogation of method and circumstance that contrasts powerfully with more neatly packaged representations.
5.3 Lines of response to quality issues
One of the trends identified in section 4.2 was an increase in longitudinal studies, and in their review Ortega & Iberri-Shea (Reference Ortega and Iberri-Shea2005) note that ‘studies adopting a case study or ethnographic approach end up reporting on only a small subset of the richer longitudinal data’ (p. 36). While this might, as the authors imply, reflect a failure of proper focus, it is conceivable that journal word limits make fuller descriptions problematic.
Since it is unreasonable to expect journals to expand in size or reduce the number of papers they publish in order to accommodate longer studies, other responses are needed. One option would be to allow space for a longer paper in one or two issues each year, which would reduce the impact on the number of papers published. Another would be to provide links to online data sets (journals such as this one and Applied Linguistics already have associated websites) or exploit the potential for electronic publishing. Markee & Stansell (Reference Markee and Stansell2007), for example, make a convincing case for the electronic publishing of conversation analytic research.
Researcher training also makes an important contribution to improving quality, and the expansion of research methods literature, courses and workshops suggests that this aspect is being addressed. A gatekeeping encounter that faces all seriously committed researchers sooner or later is the doctoral research proposal, and Kilbourn's (Reference Kilbourn2006) discussion of the qualitative proposal provides an outstanding introduction to this (a significantly briefer account can be found in the ‘top ten tips’ provided by Patton Reference Patton2002: 33–35). More neglected, but arguably even more important, is the researcher's journal/log, for which Borg (Reference Borg2001b) develops a compelling case.
Although good practice may be disseminated on a local level, there is probably a need for a more global perspective, covering publications across a range of specialities. For example, Hanke's (Reference Hanke2000) use of drawing completion as a data source when researching young children is likely to interest researchers outside the field of reading, though its source makes it unlikely that they would happen upon it. Finally, we need more first-hand accounts from researchers of their engagement with the research process, not only experiences with different elements in the research process of the sort collected by Darlington & Scott (Reference Darlington and Scott2002), but also accounts from novices (e.g. O'Toole Reference O'Toole and Walford2002) and developmental descriptions such as Giske & Artinian's (Reference Giske and Artinian2007) on using grounded theory.
It is also essential to maintain healthy debate on broader research issues. The relationship between research and practice, for example, is a fundamental topic which takes many forms. These include the relevance of research to practice (Hammersley Reference Hammersley2005), the nature of research collaboration and the teacher-as-researcher (e.g. Hawkins & Legler Reference Hawkins and Legler2004; O'Connor & Sharkey Reference O'Connor and Sharkey2004; Stewart Reference Stewart2006), the place of research in teacher education (Jones Reference Jones2004), challenges to teacher research (Borg Reference Borg2003b, Reference Borg2006, Reference Borg2007; Allison & Carey Reference Allison and Carey2007) and reconceptualisations of practitioner research (Allwright Reference Allwright2003).
6. Researcher resources
Researcher resources for QR continue to proliferate and this section is designed as a guide to some of the most useful. It focuses on books on QR methodology but also includes a brief note on journals. Comments are necessarily brief and coverage is selective.
6.1 Research in general
Not too long ago researchers in our field had to rely on general treatments, some of which are still widely used (e.g. Cohen, Manion & Morrison Reference Cohen, Manion and Morrison2007), but the appearance of books directed towards research in language teaching (e.g. Nunan Reference Nunan1992; McDonough & McDonough Reference McDonough and McDonough1997) changed this situation and it is now possible to refer to specialist texts. The challenge here is to provide wide coverage without sacrificing depth of engagement and in this respect Dörnyei (Reference Dörnyei2007) is particularly welcome. The structure of the book, moving from a consideration of key issues, through data collection and analysis to reporting, provides a solid framework in which quantitative and qualitative approaches receive balanced treatment, with a judicious mix of practical summaries, clear advice and insights from personal experience. Bulleted lists are particularly well-used, ranging from descriptive summaries to evaluative responses to key questions, and the range of perspectives is impressive. A useful complement to Dörnyei is Brown & Rodgers (Reference Brown and Rogers2002), which includes helpful exercises (and answers). If there is a primer of research in our field, this is it. More substantially, Hinkel (Reference Hinkel2005) is an essential resource, a wide-ranging collection of key papers judiciously selected to provide in-depth coverage of research in our field. Research methods take up less than 10% of the book, but the contextualisation provided by other sections is invaluable.
Perry (Reference Perry2005) writes for those who wish to make use of research in applied linguistics. The target audience is novice researchers at Master's level, but it has wider application and would appeal to teachers who wish to explore the ever-expanding research literature in our area. Though more attention is given to quantitative than to qualitative research, its chapters on locating research and reading research articles are particularly useful. The discussion in the chapter on understanding research design, taken together with the brief opening chapter, would be very useful in helping novice researchers to understand some of the conceptual issues that bedevil those new to the field. The brief guides to terminology are useful additions and the table of journals in the field provides an invaluable overview, though the index is disappointingly thin.
6.2 Qualitative research in general
The choice of general books ranges from authoritative collections designed for experienced researchers (e.g. Denzin & Lincoln Reference Denzin and Lincoln2005) to more user-friendly approaches (e.g. Mason Reference Mason2002; Patton Reference Patton2002), while the qualitative dimension in some more general treatments also makes them worthy of consideration (e.g. Bryman Reference Bryman2004). Seale et al. (Reference Seale, Gobo, Gubrium and Silverman2007a) offers a useful compromise that combines the practical orientation that a single author can provide with the range of expertise that a collection can assemble. This is a weighty collection priced for individual use and is an excellent resource for tutors. New research students will find Silverman's (Reference Silverman2004) introduction to QR particularly valuable while more advanced students might prefer his more narrowly focused guide to the research process (Silverman Reference Silverman2006).
Qualitative research in language teaching is directly addressed in two books. Richards (Reference Richards2003) offers three levels of access, from novice to (post-)doctoral, and is designed to be read either across a single level or developmentally from level to level. Its core chapters provide detailed advice on observation, interviewing and using recorded data, and the book is well-served in terms of illustrative cases. Judged by its title, Holliday (Reference Holliday2007) may seem to be a more general text, but the author writes ‘as an applied linguist’ (p. xii) and this is apparent throughout. Central to the book is the integration of writing samples, from undergraduate to post-doctoral, with a genuinely international range. The voices of researchers emerge clearly and there is also a strong sense of the author as researcher. The book is attractively written and is particularly impressive in the way it addresses core problems for novice researchers. The chapter on what counts as data addresses a fundamental but often neglected issue in a very accessible way and there is a strong sense of the visual throughout the book, both features that are missing from Richards. Holliday describes getting from data to the written product as ‘traumatic’ and his book responds very effectively to the complexities of this challenge. The very different approaches adopted in these two books produce a complementary pair with virtually no overlap – a reflection of the breadth of QR.
6.3 Ethnography and fieldwork
The standard work here (Atkinson et al. Reference Atkinson, Delamont, Coffey, Lofland and Lofland2007) is really only for the serious ethnographer, but it is nevertheless an essential collection. A more accessible resource for the novice researcher, methodologically informative, engagingly written and with entertaining examples, is Hammersley & Atkinson (Reference Hammersley and Atkinson2007). O'Reilly (Reference O'Reilly2004) is also methodologically focused and makes good use of student work. Emerson, Fretz & Shaw (Reference Emerson, Fretz, Shaw and Atkinson2001) provide an excellent brief introduction to participant observation and Delamont (Reference Delamont2002) is also worth reading for its educational focus. Bowern's (Reference Bowern2007) useful addition to fieldwork literature is aimed very much at linguists rather than language teachers, but it includes plenty of practical advice and would be an essential resource for anyone interested in researching aspects of language (phonology, syntax, etc.) in the field.
6.4 Interviewing
The standard work here is Gubrium & Holstein (Reference Gubrium and Holstein2002), but it is not designed for the novice researcher. The most straightforward introduction is probably Gillham (Reference Gillham2005), which is very accessible and covers a range of issues as well as a wide selection of interview types. However, it is essentially introductory and Kvale (Reference Kvale2008) or Arksey & Knight (Reference Arksey and Knight1999) provide deeper treatments. Wengraf (Reference Wengraf2001) offers a comprehensive guide to life history interviewing, though the book is rather hard going; Siedman's (Reference Siedman2006) phenomenological perspective is more accessible. Finally, for anyone wanting a sense of the experience of qualitative interviewing, Rubin & Rubin's (Reference Rubin and Rubin1995) personal recollections make entertaining reading.
6.5 Data analysis
This area is relatively under-represented. Miles & Huberman (Reference Miles and Huberman1994) is in some respects a useful sourcebook and Coffey & Atkinson (Reference Coffey and Atkinson1996) provide a readable discussion of relevant issues, but neither book serves as a practical guide. Grbich (Reference Grbich2007) offers good breadth of coverage and has helpful further reading sections, but the level of discussion is necessarily introductory. As a practical guide, Richards (Reference Richards2005) has much to offer. Part of the reason for the dearth of practical guides may be the growing popularity of computer software. Both Gibbs (Reference Gibbs2002) and Bazeley (Reference Bazeley2007), for example, provide excellent guides to the use of the qualitative analysis programme, NVivo, though in purely practical terms the spiral binding of the former make it much easier to use when working at a computer.
6.6 Ethics
The importance and complexities of this aspect of research are now properly appreciated, and Oliver (Reference Oliver2003) provides an excellent introduction. The book's division into two parts, stages in the research process and themes, enables the reader to select an appropriate starting point and work from there. De Laine's (Reference de Laine2000) treatment is more uneven, but the coverage is good and the writing is clear. A useful supplement to these would be Mauther et al. (Reference Mauther, Birch, Jessop and Miller2002), which offers a stimulating collection of papers engaging with the practical complexities of the field.
6.7 Classroom research
The association between classroom research and QR has always been close and a number of earlier resources are still relevant (e.g. Allwright Reference Allwright1988; van Lier Reference van Lier1988; Allwright & Bailey Reference Allwright and Bailey1991). The most wide-ranging of current treatments is probably McKay (Reference McKay2005), which would be a useful starting point for teachers interested in exploring options in classroom research, Hall & Verplaetse (Reference Hall and Verplaetse2000) making an excellent accompaniment. Anyone wishing to investigate classroom discourse will find Walsh (Reference Walsh2006a) very useful, not least because of the classroom extracts it contains. This also applies to Seedhouse's (Reference Seedhouse2004) book on the architecture of the language classroom. Even though it is not primarily intended as a research guide, its treatment of conversation analysis from the perspective of the classroom and the quality of its analysis make it essential reading for anyone working in this area. For an action research perspective, the best introduction for language teachers is still Burns (Reference Burns1999), directly addressing the interested professional and containing excellent advice. Carr & Kemmis (Reference Carr and Kemmis1986) remains a standard work on critical action research, while McNiff & Whitehead (Reference McNiff and Whitehead2005) provide a useful general guide, Reason & Bradbury (Reference Reason and Bradbury2007) a sourcebook and Edge (Reference Edge2001) a key collection of practitioner accounts. For a more detailed review of the literature, see Burns (Reference Burns2005).
6.8 Other methods
One of the most notable developments in QR since 2000 has been the growth of introspective approaches, and here Gass & Mackey (Reference Gass and Mackey2000) provide a standard introduction, including a useful table summarising relevant research. At the core of the book is a strong methodology chapter with useful advice and illustrations, though it is fair to say that the dominant orientation is towards quantitative analysis, with QR seen as having a supportive role – an important corrective to the assumption that introspective approaches can comfortably be categorised as qualitative. The authors' later book (Gass & Mackey Reference Gass and Mackey2007) on elicitation methods is more broadly based and is particularly strong in terms of the number of boxed summaries of research papers it includes. The emphasis, though, is on elicitation rather than analysis and readers interested in gaining an insight into the research process should find the discussion in Johnson (Reference Johnson2002) informative. Gass & Mackey (Reference Gass and Mackey2007) include brief sections on diary studies in two of their chapters, but anyone wishing to pursue this line should consult Alaszewski (Reference Alaszewski2006), which includes a very useful chapter on analysis.
Case studies also feature prominently in language teaching research and Duff (Reference Duff2008) provides a welcome introduction to work in applied linguistics. The book includes a particularly good discussion of the nature of case studies and a justification for them, as well as a helpful overview of research using this approach, though the book's illustrations are disappointing. From a methodological perspective, it would be useful to supplement this with Yin (Reference Yin2003), while Gomm, Hammersley & Foster (Reference Gomm, Hammersley and Foster2000) covers some of the more general and theoretical issues. Narrative research is less well served, but Johnson & Golombek's (Reference Johnson and Golombek2002) collection of language teacher narratives demonstrates what this method can offer in terms of personal investment and insight. However, the book is not designed as a guide to the method, so it is best read in conjunction with Clandinin & Connelly (Reference Clandinin and Connelly2000), which though less engaging does provide theoretical and methodological perspectives.
6.9 Other aspects
The rise of visual methods (the use of photographs, film, etc.) in QR has not yet made a significant impact in research in language teaching, though Holliday (Reference Holliday2007) may signal the beginning of a change. The same applies to internet research methods, and here Hewson et al. (Reference Hewson, Yule, Laurent and Vogel2003) is useful. It is accessible to the uninitiated and would need to be used selectively, but as far as I can judge (I have not conducted research using this method) its advice seems well-informed and practical. The case studies at the end include useful critical evaluations.
Those interested in different aspects of language teaching and learning will also find some publications that combine field specificity with research sensitivity. Hall's (Reference Hall2003) introduction to researching language and culture provides an outstanding example of this, its three chapters on research issues providing an ideal springboard for further exploration. The two chapters on interpretivist procedures and analysis that Lynch (Reference Lynch2003) includes in his book on evaluation and language assessment also make a valuable contribution to the field. Finally, for experienced researchers wishing to explore a new perspective on the research process, ten Have's (Reference Have2004) ethnomethodological approach to research methods is refreshing and thought-provoking.
6.10 Journals
The core journal for QR methods is Qualitative Research, which is very accessible and manages to avoid technical and theoretical extremes. Anyone interested in more radical and varied approaches could try Qualitative Inquiry, though papers of practical value to our field are relatively rare. Closer to home, the most convenient source is probably TESOL Quarterly's ‘Forum’, where brief contributions allow for a genuine sense of debate. A list of links to online journals, along with details of useful websites and lists, is in the appendix to this paper.
7. Conclusion: prospects and challenges
From a language teaching perspective, the first seven years of the new millennium have been a period of valuable consolidation in QR but with signs of new avenues opening up. In this concluding section I summarise some key outcomes of this review and indicate what these presage for the next seven years.
The broader field of QR has been characterised by a less confrontational orientation, with attention shifting to practical issues and away from more theoretical debates. This is a welcome development in our field, where novice researchers still become easily enmeshed in paradigmatic conundrums, and it would be reassuring to see a pragmatic approach gaining ground. If this produces more emphasis on research practice combined with greater tolerance of alternative positions, the seemingly intractable problem of defining different research traditions will be less pressing.
The most significant movement to emerge from QR generally is a shift towards mixed methods research (see Tashakkori & Teddlie Reference Tashakkori and Teddlie1998 for a useful introduction). The publication of a new journal, Journal of Mixed Methods Research, in 2007 reflects its growing importance, and research of this kind is bound to feature much more prominently in our field in the future.
This review has revealed no evidence of a continuing expansion of QR papers published in leading journals in our field, but the new millennium has seen consolidation to a point where its position seems secure. There are signs of growth in some journals, though this may now be reaching its peak. More interesting are the areas where QR is making significant contributions or opening up new territory. For example, research on CLT (Communicative Language Teaching) still features prominently and is producing interesting but contradictory local studies of how this is being applied, suggesting that this will remain an attractive area for some time to come. It is also likely that the excellent work being done in the area of writing will continue, though research in the areas of reading and especially listening is disappointing by comparison. There is surely scope here and more generally for exploring what mixed methods research has to offer, and there is a pressing need for more longitudinal studies.
Inevitably, research in language teaching has reflected wider developments, so traditions such as conversation analysis have gathered strength and topics such as identity have emerged into relative prominence. The latter is opening up new territory, generating a greater understanding of both teacher and student identity and throwing new light on the experiences of immigrants and sojourners. This has exposed unexplored features of the language teaching landscape which should attract researchers using narrative and life history methods. Linguistic ethnography is also likely to be a significant contributor to what was once seen as the hinterland of language teaching. This approach has already produced illuminating studies of the relationship between classroom and community and offers an energising reconfiguration of practical and conceptual orientations.
Mindful of the way that conversation analysis established its place in the field of L2 acquisition studies in the face of initial opposition, to the point where debates about its legitimacy are now redundant, we need to be sensitive to other emergent areas. The sociocultural dimension of engagements with language has developed immeasurably since 2000 and the prospects for deepening this knowledge are exciting.
The most encouraging development since 2000 has surely been the contribution that leading researchers in our field have made to the development of quality in QR, with the emergence of clear guidelines on standards and books designed to help the novice researcher respond to the methodological and representational demands of good research. However, on the evidence of papers reviewed here, there is still work to be done to encourage yet deeper engagement with methodological issues, especially where interviews are concerned. We need to have more details of methodological and especially analytical matters in published papers, and it would be satisfying to see the demise of summaries amounting to no more than a couple of sentences or a short paragraph.
Developments here will depend at least in part on how the Internet is exploited. The growth of online journals, tutorials, lectures, etc. will doubtless continue, but publishers could be encouraged to provide links to authors' data that will allow greater transparency. Data extracts presented in published papers can then be regarded not as windows looking out at selected features of the landscape but as doorways inviting exploration. This should encourage the sort of creative use of data that Holliday (Reference Holliday2004b) has called for, which will be realised in richer forms of representation, perhaps leading to linked data sets evolving over time, so that new forms of longitudinal studies evolve.
That said, I am grateful to an anonymous reviewer for pointing out conceptual and practical problems associated with this. For example, since QR does not treat data as autonomous or ‘objective’, their appropriation by a third party who may analyse them using a different theoretical framework raises fundamental issues about the nature of research and its representation. More seriously, if we are to share data in this way, we have an ethical responsibility to our participants to ensure that they are fully informed about – and understand – the implications of this. Neither of these problems is insurmountable, but they are potent reminders of the intellectual and ethical demands that QR makes.
Whatever the developments on this and other fronts, the relationship between teaching and research will always remain problematic, and so it should: the issues are too important to be dismissed with pat solutions. The emergence of exploratory practice as an alternative approach to action research, for example, raises interesting questions about the methodological boundaries between research and exploration. In a field where most researchers begin as teachers, the place of such ‘alternatives’ needs to be debated.
However the future unfolds, a review of this sort should conclude with a reminder of what is at the very heart of our enterprise, a poignant elision of teaching and research summed up in Freeman et al.'s (Reference Freeman, deMarrais, Preissle, Roulston and Pierre2007: 30) ‘guiding question’: ‘How can we best listen to, work with, and represent the people our work is intended to serve?’
Acknowledgements
I am very grateful to the anonymous reviewers of this paper for their close reading of a long text and their valuable comments and suggestions. Their critically supportive responses were a powerful reflection of the collegial spirit that has driven qualitative research forward so productively.
Appendix: Online resources
online journals
Asian EFL Journal
http://www.asian-efl-journal.com/index.php
A good range of papers, some of them qualitative in orientation, but discussion of research methodology does not feature strongly.
Educational Research and Reviews
http://www.academicjournals.org/ERR/Archive.htm
Wide coverage and only one QR paper on language teaching so far, but good for local educational contexts.
Education-line
http://www.leeds.ac.uk/educol/
So far, very little in this area, but access to conference papers and work-in-progress makes it a potentially very valuable resource.
Educational Researcher
http://www.aera.net/publications/?id=317
Some useful papers on general issues in research. More likely to be of interest to the teacher of qualitative research methods than students.
English Teaching Practice and Critique
http://education.waikato.ac.nz/research/journal/index.php?id=1
This focuses predominantly on L1 contexts and literacy issues, but there are some useful papers relating to the research process, for example the December 2000 issue on what counts as research on English/literacy issues.
Forum: Qualitative Social Research
http://qualitative-research.net/fqs/fqs-eng.htm
A key website for the qualitative researcher.
International Journal of Qualitative Methods
http://ejournals.library.ualberta.ca/index.php/IJQM/index
This goes back only as far as January 2007 but has already published some papers that are likely to be very valuable to the teacher of qualitative methods. Practitioners' accounts are particularly revealing.)
Journal of Research Practice
http://jrp.icaap.org/index.php/jrp/index
This started in 2005 and already includes some useful papers on aspects of QR.
Prospect
http://www.ameprc.mq.edu.au/resources/prospect
A leading journal in the field and included in list of journals analysed in the state-of-the-art paper.
The Qualitative Report
http://www.nova.edu/ssss/QR/index.html
Another key website with useful articles and excellent links. Some useful discussion of the thinking behind it is found at http://www.nova.edu/ssss/QR/QR12--1/chenail.pdf and would be useful to anyone contemplating setting up an online journal.
Qualitative Research Journal
http://www.latrobe.edu.au/aqr/index.php?option=content&task=view&id=22&Itemid=41
The latest issue is not available but archived issues are and contain useful papers on aspects of QR methodology. However, you have to download the complete issue (around 3MB) if you wish to read a paper offline.
TESL-EJ
http://tesl-ej.org/ej41/toc.html
This is a good source of qualitative studies, many with useful discussions of methodological issues. Issues appearing over the last four years are more productive in this respect than earlier ones.
websites
Association for Qualitative Research (Australia)
http://www.latrobe.edu.au/aqr/
Useful links to conferences and the online journal, though at the time of writing the resource page was still under construction.
The Grounded Theory Forum
http://www.groundedtheory.com/
Relatively limited resources for non-members, but some useful references.
Social Research Update
A list of excellent brief introductions to aspects of research.
ESRC National Centre for Research Methods
http://www.ncrm.ac.uk/research/outputs/publications/
Excellent resource with very useful methodology papers available online.
Qualiti – QR Methods in the Social Sciences: Innovation, Integration and Impact
http://www.cardiff.ac.uk/socsi/qualiti/
A good source of information and access to working papers and ‘Qualitative Researcher’
National Science Foundation Directorate for Social, Behavioral and Economic Sciences
http://www.nsf.gov/dir/index.jsp?dir=SBE
This covers a lot of territory and includes useful publications.
British Educational Research Association
http://www.bera.ac.uk/publications/index.php
Links to some useful publications (for purchase), useful ethical guidelines and a brief guide to good practice in research writing.
lists
Message to: mailbase@mailbase.ac.uk:
Message: Join BIOG-METHODS
(Focuses on biographical methods)
Ethnography-in-education@mailbase.ac.uk
Message to: mailbase@mailbase.ac.uk
Message: join ethnography-in-education [firstname lastname]
Ethnomethodology/conversation analysis
Message to: comserve@cios.org
Message: join ethno
Message to: comserve@cios.org
Message: join methods
(Designed for methods tutors)
Message to: mailing-list-request@qsr.com.au
Message: SUBSCRIBE QSR-FORUM
(Designed for users of NUD*IST and Nvivo qualitative analysis software)
Message to: listproc@scu.edu.au
Message: subscribe QUAL-L [firstname lastname]
(Qualitative research list)
Qualitative Research for the Human Sciences.
Message to: listserv@uga.cc.uga.edu.
Message: subscribe QUALRS-L.
Keith Richards is an associate professor at Warwick University. His research interests lie in the area of professional interaction and his current work involves an exploration of academic argument and identity. His recent publications include Qualitative inquiry in TESOL (Palgrave, 2003), Applying Conversation Analysis (edited with Paul Seedhouse; Palgrave, 2005), Language and professional identity (Palgrave, 2006) and Encounters in TESOL: Discourses of teachers in teaching (edited with Sue Garton; Palgrave, 2008).
