The Craft of Knowledge

Experiences of Living with Data

C. Smart, J. Hockey, A. James, C. Smart, J. Hockey, A. James

• Publisher: Palgrave Macmillan
• Year: 2014
• Print ISBN: 9781137287335
• eBook ISBN: 9781137287342
• Topic: Social Sciences
• Subtopic: Information Management
• Index: Social Sciences

This book is a contribution to contemporary debates on social research with a unique focus on the relationship between methods and the crafting of knowledge. Nine experienced researchers from different disciplines have come together to explore what really matters to them in the process of doing qualitative research.

Part I

The Changing Politics and Context of Research

1

‘Bias Binding’: Re-Calling Creativity in Qualitative Research

Simone Abram

This chapter reflects on the methods, tools and experience that make up the practice of qualitative research to consider what data are and what data do. Referring to a range of different research projects I have undertaken or been involved with, I ask what counts as data, and where and how they intersect with experience, and in doing so I question a distinction now increasingly being made between the personal and impersonal qualities of data/experience. Noting another distinction, between methodology and methods, I explore, in particular, the relationship between methods and the idea of ‘bias’ that increasingly seems to worry students that I encounter. And, as I go on to argue, the notion of bias has revived, for me, ideas about both the necessarily embodied nature of the research process and about the ways in which creativity has to be seen as a critical part of the experience of doing qualitative research (see James, this volume).

Using research methods or doing research?

There appears to have been a wave of interest in research methods in recent decades that shows little sign of waning and a proliferation of books on methods supplies this expanding field with instruction, inspiration and advice of the kind that is often particularly reassuring to postgraduate students exploring anew the world of research. Thus, although treatises on research methods have always characterised sociological research (early sociological texts were indeed pleas for new methods, such as Durkheim (1897) arguing for statistical analysis), methods seem to be taking on a life of their own as a subject of teaching and training. Several reasons might be noted for the intensity of attention now being given to methods.

First, increasing emphasis on interdisciplinarity, and the flourishing of pluri-disciplinary studies, often leaves methods as the easiest common point of discussion between students, scholars and funders. In the multidisciplinary departments where I have taught in recent years (in planning and tourism studies), for example, postgraduate research students have widely divergent theoretical interests. When they present their research projects to each other, they are urged to focus on methodology, since this is the area where they can be expected to find some commonality. Students in a broad interdisciplinary school may have no specific interest in another student’s topic of study, but everyone can share lessons about methods.

Second, methods have also come under increasing scrutiny from funding agencies, such as the United Kingdom (UK) research councils, who have begun to include methodological innovation as a criterion for research awards. Methods appear to be an easy target for interdisciplinary competition, since innovation can appear in any discipline, from economics to anthropology. There seems to be little evidence, however, that this emphasis on new methods actually leads to radical invention, as opposed to simply transferring methods from one discipline to another (see Wiles et al. 2011) and, in the UK at least, this increased attention to methodological innovation has also signalled a shift away from substantive intellectual questions towards the demonstrable practical effects of research: research is now evaluated for innovation in the methods employed and for impact in its implementation.

Despite this interdisciplinary focus on and discussion about methods, the degree of reconciliation between methods employed in different disciplines is still limited, however. Hence we find discussions about ethnography gaining ground in diverse disciplines right across the social sciences, but with a wide diversity in what is meant by ethnography in different contexts. While there are still anthropologists who think that ethnography belongs to them, the anthropological tradition of ethnography is now one among many schools of practice, for whom participant-observation, interviewing, oral histories or mapping form a broad toolkit of methods to mix and match.

A third reason for this strong focus on methods may also be the increasing intensity of the governance of ethics, since ethics approvals are heavily geared towards the specifics of the methods being proposed, rather than the ethics of doing the research per se, or the philosophical approach adopted by the researcher. Indeed, methods are sometimes treated in ethical approval processes as an element of research that can be evaluated almost irrespective of the context or content of the research, its theoretical development or its empirical novelty.

One important consequence of this tendency has been to divorce methods from the body of the researcher who will implement the methods. By this means, methods become seen as impersonal and inherently teachable, as a set of learnable skills that will, in and of themselves, produce data, irrespective of the researcher who will implement them. Indeed, it may be in these ideas that the notion of ‘bias’, noted earlier as so troubling to students, has taken root. And yet this divorce is illusory as I go on to detail below, through three empirical examples drawn from my own research experiences. It is also the case that, ironically, researcher funding is nowadays increasingly tied to particular embodied individuals who are excellent researchers. Special personal fellowships and funds exist for ‘outstanding’ individuals, at both national and international levels (for example,. the requirement for European Research Council Advanced Grants (ERC) is for ‘exceptional research leaders only’). In this way, the person of the researcher is increasingly significant, and one might argue, even fetishised, even though, ironically, the research may in fact be carried out later, by a research assistant, rather than the researcher him or herself.

Notwithstanding such caveats, having evaluated ethical approvals or grant applications and regulations over some years, it would seem that, nowadays, the ideal research relationship for many researchers has become one of pure knowledge – one knows the field and gets the data to analyse, using appropriate methods, but one is not tied up in anything so messy as an embodied relationship with the subjects of research. This suggests a rather ambivalent dichotomy, rather than a straightforward one, between the personal and impersonal in research, and it is one of several such ambivalences that I now go on to explore.

Methods and methodology

Standard ‘how-to’ books on sociological method continue to be published alongside radical critiques of the very idea of method at all. It might be argued in this context that two quite distinct literatures on qualitative research methods have emerged in recent years. The former, as noted above, is directed towards teaching a broad range of qualitative research methods to students at undergraduate and postgraduate levels, and includes a diverse array of textbooks, many of which include instructions on how to do research, how to apply methods and how to analyse data (for example, May 1993; Silverman 1997; Flick 1998; Seale 1998).

The literature on methodology, by contrast, is a discussion among social science researchers on the meaning of method, one that challenges the epistemology of data and re-situates ontologies. Post-positivist approaches, for example, question the status of knowledge and meaning and, in particular, the relationship between spoken and embodied knowledge. John Law, for example, asserts that in professing the use of rigorous methods for accessing data, the social sciences are engaged in a kind of epistemological hygiene, unable to accept the messiness of the world and unable either to analyse or even describe it (Law 2006). He characterises advice about methods as tips for research: useful, but mostly concerned with methodological cleanliness in ways that do not work in practice. In brief, he argues that such methods-advice presumes a realist ontology, that the world out there is available for reliable documentation, if only we use the right methods correctly (Law 2004).

Of course, Law is not alone in his critique, and much theoretical discussion on the structure and practices of empirical research is to be found, not least in social anthropology, where reflexive consideration about the construction of research fields and objects is well established, having followed on from the ‘writing culture’ debate of the 1990s (see Gupta and Ferguson 1997; James et al. 1997; Stoller 1997; Amit 2000). Anthropologists are familiar with the idea that a research field is a unit imagined by the researcher, a framing mechanism for the convenience of research, and one that entails consequences, in terms of the visibility and invisibility of particular viewpoints. Anthropologists have also long been more or less obsessed with the embodied experience of fieldwork and its impact on the ethnographer, as well as the ethical dilemmas that continually present themselves in participant-observation, often considered to be ethnographers’ central (if not exclusive) method.

From this perspective, the increasingly prevalent idea noted above – that there are independent data waiting ‘out there’, to be collected by a range of accepted methods, applied to particular circumstances becomes problematic. This is a scientistic, or modernist, notion that presumes a universality and uniformity in human encounters; it suggests that data exist outside the realm of human (or human/non-human) relations – data, as it were, beyond the reach of bias. It is also an approach that presumes that, although the combination of methods to be used may vary, the way of doing these methods and the kinds of data generated are relatively standard. By implication, then, the person doing the research, once equipped with the relevant skills, must also be standard. Any researcher should be able to generate empirical data – or so it is assumed.

Yet many anthropologists have argued that this is not the case and, as Okely points out, Malinowski – the pioneer of embodied research – noted in his fieldwork diaries of the 1910s that ‘theory creates fact’ (Okely 1996: 38). It is important to acknowledge, then, that the work of objectivising data, via an increasing focus on method, has largely been a project of the mid-twentieth century, and has included a great deal of forgetting of what was well known by our predecessors.

In teaching research methods to students from diverse disciplinary backgrounds over the last decade, however, it has become increasingly apparent to me that these two approaches – methods versus methodology – sow confusion among students. This is particularly the case for postgraduate students who have not come from an anthropological background. For them, the ‘standard’ approach is very comfortable, while methodological discussions relating to the epistemology of research represent an existential challenge. Doubt cast over the status of data translates quickly into a doubt over the status of fact, and for those students with a more scientistic education, this can be deeply disturbing.

Confusion is also engendered by the way the two literatures, indicated above, often use the same vocabulary, but imply different things (in what Richard Cowell has called ‘rhetorical isomorphism’, Abram and Cowell 2004). The example of ‘ethnography’ is particularly apt. It is used to mean anything from prolonged personal involvement in extensive participant observation (typically in social anthropology), to going to public events or meetings as a complement to structured interviewing (such as in some management research). The two literatures may also be mixed, even in the same volume, as a postmodern or post-structuralist critique of methodology gives way to a ‘how-to’ of methods based on realist assumptions. Clive Seale’s (1998) admirable volume on researching society and culture is a case in point. He does what many of us do when teaching research methods, beginning with questions about philosophies, histories and theories of research and the status of knowledge, before advising how to devise questions for a social survey and how to treat them statistically. On the one hand, he is pointing to the socially embedded nature of qualitative social research, and then, on the other, treating it as a source of empirical and countable data – data without bias.

Responses from students illustrate this contradictory situation. I have frequently encountered students who ask me how they should analyse their interview data and draw conclusions from it but, in my mind, this question starts from the wrong end of a project. How can one begin research without a theoretical context and a disciplinary debate through which to interpret research experiences, whether we call that data or not? Merely to ask the question ‘how should I analyse interviews’ implies that this is a mechanical process akin to applying a chi-square analysis to a statistical dataset. This reveals a very particular understanding of qualitative research and of the data they produce that is radically different from an anthropological approach that makes its focus the more intuitive and personal nature of the data it seeks to analyse. What students often fail to grasp is that while statistical methods offer a range of formulaic methods to manipulate disembodied data, qualitative methods are far less amenable to externalisation precisely because much of the data remain largely embodied.

Further, the question raised by students about analysing data suggests that it is possible to create transferable knowledge for the application of the intellect to the case at hand. Organising material through coding or indexing is one way to help do that task, and this is what can be taught. Yet what characterises inspirational or original research is precisely the application of theoretical concepts to diverse materials and the exercise of creative inspiration, as James (2013) persuasively argues. Indeed what distinguishes inspiring from mundane research is intellectual creativity. So, the implications of the question raised by students are not merely about the virtues, or otherwise, of specifying methods for analysing data. More significantly they deny the role of creativity in research.

But, it is this very creativity that raises a further question for many students: how to avoid ‘bias’ in their research projects. One can only speculate (in an informed way) on the predominance of scientific paradigms in school science education, the general dominance of positivist models in university undergraduate courses or the low profile of social sciences in contemporary society that underpin this question. However, the very persistence of bias-talk suggests that we can no longer make presumptions about the broader acceptance of social science debates on knowledge or methods in relation to methods teaching.¹

This tension between the idea of impersonal research skills and the personalised experiential research career noted above is exacerbated by a third factor: the nature of university careers. Promising young researchers complete their doctorates, perhaps conduct a post-doc research project as a full-time researcher, with some degree of teaching in addition, in order to advance their research skills and research profile. If the scholar moves into a permanent university position, they are most likely to start as a lecturer, at which point the teaching and administrative load often precludes the possibility of doing hands-on empirical research and certainly anything like extended ethnographic fieldwork. In many university departments, as an academic progresses through his or her career, teaching and administrative loads, as well as grant-development demands, mean that they are strongly encouraged to employ research assistants (usually the post-docs noted above) to carry out any empirical work. The first job to go is usually the transcription of interviews. Yet, as most fieldworkers know, this is the process through which you become intimately familiar with interview data. The next thing to be handed over for the research assistant to do is the interviewing, and so it goes on.

Fieldwork is thus relegated, in many disciplines, to a kind of manual labour task, a practical detail that can be satisfied by junior researchers, leaving senior academics to manage research projects and publish results. Thus, the experienced researcher who has spent several years honing their fieldwork skills is, in practice, rarely able to use them. This classic bureaucratic trajectory, in which able practitioners are gradually siphoned off into management roles, fails to acknowledge, the degree of expertise, sensitivity and skill required of practical research work and the intimate links between experience and knowledge that I have indicated above. The dangers, therefore, in seeing research methods as a means of simply producing data are twofold. First, intellectual creativity is strained out of the very personal and professional experience that goes into exciting and original research. Second, the teaching of disembodied methods perpetuates the idea that ‘bias’ is not only undesirable but also avoidable, rather than an inevitable element of all research that can be acknowledged, reflected upon and incorporated into a considered methodological approach.

Thus, methods, I am arguing, offer us guidelines and suggestions; but we must adapt and adjust them into a methodology, through reflection on the context of their application and with judgement about their efficacy in leading towards particular kinds of outcomes. With so many circumstances conspiring against us, then, how can we reclaim ground for the role of the empirical researcher as intellectual actor? How can we illustrate the centrality of embodied experience and reflection in qualitative research, and strengthen, rather than question, the legitimacy of the creative link between practical research and intellectual activities?

Fortunately, there are some allies. In anthropology, for example, the role of particular researchers and writers has long been recognised as central to the history of the discipline, at least since the publication of Adam Kuper’s Anthropology and Anthropologists (1973). Elsewhere, Somekh (2003) has adopted Michael Polanyi’s 1958 notion that passion can be a guide to discovery, recognising the centrality of interpretation of theories and ideas in the research process. She also mentions the idea of ‘phronesis’ or practical wisdom, which Flyvbjerg (2001) has taken up as a concept to understand the knowledge of professional practitioners. Somekh writes a portrait of a distinguished educational researcher, John Elliot, who insisted that a teacher could generate theory by ‘pitting his own intellect and creativity against the theories and ideas of another researcher or philosopher who provides him with insights’ (2003: 255; see Rapport this volume). Indeed, the idea of imagining oneself in a debate with other theorists is one that I emphasise to research students as they learn how to write for academic publication as well as for popular dissemination. It is one that emphasises the personal and positioned role of each res...

Table of contents

• Cover
• Title
• Copyright
• Contents
• List of Figures
• Foreword
• Notes on Contributors
• Introduction
• Part I The Changing Politics and Context of Research
• Part II Research and the Crafting of Knowledge
• Part III Living with Data
• Bibliography
• Index