Methodological innovation and research ethics: forces in tension or forces in harmony?

Introduction and background

Methodological innovation and research ethics have each been the subject of increasing interest within the social sciences, keenly watched over by powerful research councils and the subject of journal special issues. Yet, the relationship between the two – innovation and ethics – has rarely received attention. There are potential tensions between them, yet the reality is more complex, and this article looks at how ethical issues are a part of innovation. This interest in the innovation–ethics dynamic did not start out as the focus of our research but instead emerged from discussion of the data from our case studies of methodological innovation in social science research. The discussion was, in part, stimulated by the fundamental and unavoidable question of whether methodological innovation is inherently a ‘good’ thing. Hammersley (2008) is explicit that he does not equate radical research with good research. ‘Good’ research methods can be understood as methods that are able to address important social research questions in ethical ways – thus, their virtue is multilayered and situated. In this formulation, methodological innovation might be seen as associated with some kind of beneficence agenda (Rhodes, 2010): doing good for the research community. We acknowledge the unavoidable morally charged nature of judgements of goodness applied to research quality or ethics. Indeed, the whole relationship between being innovative and being a good and responsible researcher is complex, as ethics is more than the avoidance of harm prioritised in the regulatory approach, but a balance of risk, efficacy, justice and respect (Rhodes, 2010) and promotion of integrity, quality and transparency (Economic and Social Research Council (ESRC), 2010a). This article engages with the desire to be a responsible and ethical researcher as one of the less frequently discussed drivers of innovation and addresses the perhaps inevitable tension of research ethics both driving and constraining innovation in research methods and practices.

As Coffey (2010) observes, the growth of interest in innovation in social research methods reflects the way it has been regarded (in the United Kingdom at least) as necessary to sustainability. The ESRC is a powerful UK government funding body for social science; innovation is part of its agenda as evidenced by its investment in the National Centre for Research Methods with its brief to identify and foster methodological innovation. Yet, innovation in government and research council policy terms remains somewhat unformulated, and ill-defined and abstract discussion of the phenomenon is limited in usefulness.

There are researchers who question the concept of innovation. Hammersley (2008), for example, argues that the radicalism in qualitative research emerging now, rather than being novel, is a return to earlier radical orientations. Travers (2009) argues that there are organisational and cultural pressures encouraging researchers to demonstrate and pursue innovation but that many developments presented as innovations are largely ‘fads’, which fail to adequately address long-standing methodological challenges. Many developments claimed as methods, Travers argues, are transient and lack the considered process of development of more established methods, such as the Chicago School tradition.

This critical stance on the whole concept of methodological innovation is a counter-narrative to an emerging literature on who innovates and why, how their work becomes recognised as innovative and who adopts new practices (see Kozinets, 2012; Pain, 2009; Travers, 2009; Wiles et al., 2011; Xenitidou and Gilbert, 2012). Xenitidou and Gilbert (2012) have concluded that innovative methodologies: ‘primarily entail crossing disciplinary boundaries’, ‘usually entail the use of existing theoretical approaches and methods in reformed or mixed and applied ways’, ‘entail the use of technological innovation’ (p. 2) and reside both inside and outside traditional academic institutions. For commentators such as Hesse-Biber and Leavy (2008), new methods have emerged amid new questions and insights. Phillips and Shaw (2011) illustrate this in their reflections on how the ‘call for innovations’ (p. 609) in social work research has come at a time of social and political shifts. They highlight not only the importance of reflecting on how methods are impacted by such shifts but also the need for ‘a moral response’ in upholding ‘core social work values’: ‘Methodological choices, indeed all aspects of research practices, are not innocent and can rupture or contribute to the negative effects of societal changes’ (p. 610).

Controversy and some lack of conceptual clarity abound in research ethics similarly. Sikes and Piper (2010), for instance, argue that despite increased interest in research ethics, the literature is dominated by ‘meta-ethical overview’ (p. 205). There is consensus, though, that the landscape of research ethics has changed fundamentally as governance has played a stronger role, and gatekeeping processes have become formalised and bureaucratised. This process has been referred to as ‘ethics creep’ and has been widely criticised by UK social scientists (Dingwall, 2008; Hammersley, 2009) as well as social scientists in other countries. Schrag (2010) plots this history in the United States, and Van Den Hoonard (2011) examines this phenomenon in Canada while also providing an international overview. Israel and Hay (2006), when pondering the ‘division, mistrust and antagonism’ between ethics regulators and social scientists, argue this is both disturbing and ironic given their shared starting point ‘that ethics matter’ (p. 1). The concerns raised are that increasing ethical regulation encourages the avoidance of risk and that this threatens the future of innovative research. It has been identified as posing particular problems for ethnographers (Murphy and Dingwall, 2007), visual and creative researchers (Prosser and Loxley, 2008) and online researchers (Orton-Johnson, 2010). Alongside critics of the system, the importance of researchers engaging positively with systems of ethical review has also been identified (Iphofen, 2009; Israel and Hay, 2006). We return to the role of ethics review later in this article.

Methods

In this study, we examined three particular cases of innovation as a means of focusing on the nuanced detail of methodological innovation in action. Our approach was to investigate ‘a contemporary phenomenon within its real life context; when the boundaries between phenomenon and context are not clearly evident’ (Yin, 1989: 23). The cultural and temporal context included controversy around the concept of, the need for and the nature of innovation. It comprised ‘a rapidly changing and globalizing world, amidst social progress and change, as well as theoretical developments in multiple traditions both within and across disciplines [where] new research questions are being posed or re-examined’ (Hesse-Biber and Leavy, 2008: 1). Our aim was to explore, from multiple perspectives, why the particular methods were developed. Moreover, we aimed to examine how key adopters, developers and advocates of, or commentators on, the methods regarded ownership, adoption and adaptations to the methods. Using the data, we additionally sought to gain theoretical insights (Bassey, 1999) and ‘unique and universal understanding’ (Simons, 1996: 225) of the interaction between methodological innovation and ethics.

Previous research had shown that what is claimed as ‘innovative’ often relates to adaptations to existing methods or to the transfer and adaptation of methods from other disciplines (Phillips and Shaw, 2011; Wiles et al., 2011); therefore, the task of selecting cases of innovation was far from straightforward. Indeed, we know that the roots of the identified innovations do not lie with the named innovators; rather, these people have some claim to a critical juncture in their emergence. Ultimately, cases were selected that (a) had been in existence in some form for around 10 years, to allow time for ‘take-up’ by the wider social science community; (b) had been identified as ‘innovations’ through a narrative review (Pain, 2009; Wiles et al., 2011) or other research on innovation (e.g. Xenitidou and Gilbert, 2009) and (c) were qualitative methods. While critics of the concept of methodological innovation might see the innovative aspects as exaggerated, the three cases were chosen as exemplars of innovations, addressing methodological challenges in terms of enabling the study of a new area of social life, providing insight into the aspects of social life that are difficult to access by traditional methods or managing ethical, access or response issues raised by traditional methods or approaches. They may be filling what Hesse-Biber and Leavy (2008: 4) refer to as ‘methods gaps’.

The cases comprised the following:

We summarise these cases here before briefly describing our own research methods.

Netnography, developed by Robert Kozinets from Canada, sits within a broader methodological context of online/virtual ethnography as a way of researching online communities; it thereby represents an exemplar of methods that enable researchers to focus on a new area of social life. Kozinets’ discipline is marketing, and his interest in web discussion groups is to inform the business community regarding consumer choices and thinking. There is no one accepted way that online communities should be studied, and while online ethnography involving observation of naturally occurring ‘postings’ and ‘threads’ within an online forum is common, data collection offline as well as online may be conducted (Hine, 2000). Following the rapid developments in the online world, there has been burgeoning interest in online research methods, including online interviews and the analysis of material from websites, chat rooms, blogs and other forms of social media. Hine’s (2000) ‘Virtual Ethnography’ marked development of methods in this field among a range of publications on the topic from the late 1990s onwards; it is included as one of Hesse-Biber and Leavy’s (2008) emergent methods (Hine, 2008). In Wiles et al.’s (2011) review of claims to innovation, online and e-research methods represented the fourth largest group of innovations among the articles identified. Xenitidou and Gilbert (2009) identified ‘netnography’ as a form of online ethnography representing an innovation. Kozinets developed his ‘netnographic’ approach to online research within the relatively new disciplinary field of Marketing and Consumer Research. In netnography, he adapts traditional ethnographic research techniques to the study of cultures and communities that emerge though computer-mediated communication, and presents this as a new, qualitative, economical, effective and unobtrusive means of studying ‘naturally occurring’ online communication and behaviour, and generating naturalistic data about online communities (Kozinets, 2002, 2010). Kozinets’ particular contribution is a pragmatic, applied and systematised approach intended to address many of the procedural, ethical and methodological issues specific to online research.

Child-led research is an approach pioneered by Mary Kellett at The Open University in England. Kellett is an educationalist concerned with children’s development and empowerment. The innovation involves providing training and support to children to enable them to design and conduct their own research. Like netnography, it is a development of its time, in this case, emerging against the backdrop of the new sociology of childhood and moral and ethical standpoints about the importance of children’s voice and children as social actors. Kellett (2010) argues that children need to lead research on children because of their unique ‘insider’ perspective, providing understanding of children’s worlds that are inaccessible via research led by adults. Like Kozinets, Kellett has popularised her innovation by systematising her approach and published step-by-step guidance – this time on training children in research methods (Kellett, 2005). Child-led research sits within a broader range of participatory approaches, which include user involvement, emancipatory research and partnership research (Frankham, 2009) increasingly adopted with groups of people viewed as vulnerable or socially disadvantaged. The child-led approach pioneered by Kellett represents a method that has sprung from moves towards interdisciplinarity and provides an exemplar of a method or approach developed to manage the ethical, moral and access problems that traditional methods pose. Among the innovation claims reviewed by Wiles et al. (2011), around a third of the 57 articles identified cited moral or ethical reasons for the innovation, and many of these related to issues of empowerment.

‘Creative research methods’ developed by David Gauntlett (2007) encompass a range of methods, including visual, performative and sensory. Gauntlett’s discipline is media and communications, and his focus is on digital media and identities. His creative work has focused in particular on the participant creating something (a photograph, video, drawing, scrapbook, collage or model) that is then used within the research process, usually for data elicitation purposes. He has not particularly branded his work with a sound bite name, but his innovative contribution involves the reflective process of creating a three-dimensional artefact notably a Lego model to metaphorically represent the creator’s identity. This case is an exemplar of a method that is claimed to provide insight into aspects of social life that are not accessible by traditional methods. Here, the timeliness concerns developments in visual methods more generally, which have reached a key point in research awareness (Prosser and Loxley, 2008). Among claims to innovation in qualitative research between 2000 and 2009, Wiles et al. (2011) found ‘creative methods’ to be the largest group of innovations among the articles identified. Gauntlett has been exploring the ways in which researchers can work with people’s ability to create and reflect during the process of production.

Our research centred on the above three cases of innovation. It comprised a systematic search of the literature to explore the response to the innovations within the academic community plus semi-structured interviews with the innovator, with 5 or 6 people per case who were able to comment on its usefulness and development without any requirement to prepare. Table 1 outlines the interviewees and how they are denoted in the text. In all, 3 interviews were conducted via skype, 1 by email, 10 by telephone and the remainder (3) face-to-face in a venue of the interviewee’s choosing. Interviews ranged from 30 minutes to 2 hours and were fully transcribed.

OPEN IN VIEWER

Table 1.

Interviewees for each case.

Interviewees for each case	Abbreviation
Early career researcher applying or adapting the innovation	ECR
Experienced researcher in the area familiar with the work	Exp
Book reviewer of the innovator’s work	Rev
Knowledgeable researcher/user of the innovation from a different country than that in which the innovation originated	Int
Knowledgeable researcher/user of the innovation from a different discipline	Disc

As the study put our peers at the focal point of the research, there were accompanying ethical sensitivities. We were conscious that it would be unfair to invite the participation of innovators and then expose them to public criticism, particularly as their anonymisation was not possible, and we were not giving them a right to reply. Aware of potential vulnerabilities, we provided the innovators with transcripts of their interviews and drafts of all articles for their comment. Other interviewees were given their transcripts for checking and amending, and we have done our best to anonymise these individuals. We acknowledge that while some of the criticisms made by interviewees may be inaccurate or easily defended, we have not sought such dialogue for our purposes here.

Analysis of the transcripts was conducted without the use of computer-assisted qualitative analysis software. The approach was iterative and thematic with global and organising themes emerging from our own theoretical approach and related to issues of timeliness, distinctiveness, contribution, breakthrough/acceptance/impediments to acceptance and future developments of the innovations. Within these categories, sub-themes were identified and general and specific points related to these were identified, highlighted and refined by the research team. The analytic process, which was influenced by Thematic Network Analysis (Attride-Stirling, 2001) and Framework Analysis (Richie and Spencer, 1994), culminated in a summary for each case. The summaries, thematic coding and original transcripts were then re-examined to pursue the question of the interaction between ethics and innovation.

Innovation and ethics: is innovation inherently a good thing?

The value of qualitative methodological innovation for social science research is debated in the literature. For Hesse-Biber and Leavy (2008) ‘innovation in the practice of social research is crucial’ ‘for enhancing our understanding of the human condition’ (p. 12). For Lincoln (2005), an attraction of qualitative research is ‘the promise and democratic and pluralistic ethics of qualitative practices’ with ‘a fresh cadre of methodologists committed to seeing social science used for democratic and liberalizing social purposes’ (pp. 165–166). For the ESRC (2010a), innovative research is important enough to be worthwhile despite the risk of failing to deliver the usually expected outputs and impact. Phillips and Shaw (2011), in contrast, warn against equating innovation with progress and reform in ‘an uncritical romanticisation of any research practice because of its novelty or technological prowess’ (p. 610). Nonetheless, they, as do we, also find sympathy with Denzin’s (2010) argument that addressing social justice should characterise researchers’ innovations and ethical responsibilities.

Lincoln (2005:166) postulates that ‘qualitative research may be compromised or even threatened by the new methodological conservatism being propagated in the name of evidence-based research and “scientifically based educational research”’ plus, associated with failures in biomedical research, greater scrutiny by Institutional Review Boards. While Lincoln refers to the United States, this phenomenon is recognisable, if not universal, in the United Kingdom and elsewhere (Coffey, 2010). Sikes and Piper (2010), for example, were inspired to edit a journal special issue after their own difficulties getting ethics clearance. They also cite the claim by Israel and Hay (2006) that in the United States, Canada, the United Kingdom, New Zealand and Australia, ‘social scientists are angry and frustrated’ at their ‘work being constrained and distorted by regulators of ethical research who do not understand social science research’ (p. 1). Following an ethnographic study of research ethics review, Van Den Hoonard (2011) makes a powerful (seductive!) argument that such regimes are ‘seductive’ for universities and funding bodies. Despite an acknowledged ambivalence about the ethics review process, he blames this seduction and the guardian role of review boards for a transformation in the very nature of research methods – a kind of homogenisation, whereby ‘the inherent richness invested in the diversity of social science disciplines is disappearing, leading to their pauperization’ (p. 3). Here, we see then the two interacting forces in tension: methodological development pushing forward ethical research practice and institutionalised research ethics practices pushing back methodological developments.

Among our interviewees, there was some adoption of the position that methodological innovation is a desirable phenomenon. One experienced methodologist commented, ‘of course you want innovation’ (Disc3), and another particularly valued the willingness to ‘push against’ and challenge resisting ‘brick walls’ (Exp5). Sometimes innovation was seen as good in its place (Disc3), that place being ‘a possibility’ among ‘a whole palette of possible research methods’, while critically evaluating its usefulness (Exp4). For one methodologist, there was worry about the politics of innovation, whereby newness is valued before methods have been properly evaluated – ‘about it being this big machine’ whereby ‘the whole aim is to just churn out something just because it’s new’. The argument continued, ‘I don’t think, just because it’s new, means it’s going to be any good’; indeed, ‘you might have new methods, but they might not tell you anything very new or very interesting’. Here, we see a different view on the ethical imperative to innovate: ‘instead of seeing innovation as just applying new methods … willy-nilly’, there is concern with ‘how we ask new questions and find new ways of seeing the world, and then think[ing] about how we might get at those with different kinds of methods’ (Exp4). Another response to this tension was that ‘innovation doesn’t necessarily lie in method; it lies in our reflexivity towards using the method’ (Disc2). Reflecting on our dataset of interview transcripts we lend support to this stance that innovation is as much about reflexivity as about new techniques in themselves. Indeed, we go further in this article by making this link with research ethics also; innovation in research is not so much a research virtue in and of itself but often represents a response to a desire to do something virtuous, making the innovation an ethical act.

This observation brings us to the issue of drivers for methodological innovation, which may not be transparent in writings about the method or its application but about which we asked interviewees to reflect. One respondent reflected that ‘we tend to assume that methodological change is intellectually driven; which it is, of course … but the technological affordances are very important’ (Rev1). Furthermore, technological innovations can afford researchers opportunities to work with data that are readily available, easily readable by machines and therefore that ‘grows without a lot of reflexivity’. Technological affordances, therefore, add to the mix of tensions in the interaction between ethics and innovation particularly pertinent to the case of netnography, though in conventional ethnography, non-obtrusive research and covert observation are equally possible.

In the remainder of this article, we address our main points of reflection with respect to how the desire to develop and apply a methodological innovation in qualitative research meets the desire to do research that is good in quality and/or ethically good. Thus, we discuss the way researchers treat people and the ways in which researchers act responsibly, the democratisation of research – in terms of the ethics of valuing different forms of expertise and the ethics of ‘everyone’, the issue of empowerment and the relationship of research to the academy. We conclude with consideration on the formal process of ethics peer review. These are all themes that have emerged through the analysis of our interview data.

Treatment of others

A central premise in research ethics is the ethical treatment of others, particularly participants in the research. The desire to treat people well was important to all the innovators. Kozinets was acutely aware of the potential in online research for researchers to mine data from online forums but argued that it is better research and more respectful of the people within online communities to ‘participate as a typical blog reader or a member of that community member would’, thereby following ‘communal rules’. For Kozinets, ‘participation [by the researcher in the online space] doesn’t necessarily mean interfering in some way, it means living as a culture member does’. This respectful stance was welcomed by others adopting the innovation; one interviewee noted that Kozinets makes powerful arguments about the value of this compared with data mining approaches (Rev2). Similarly, the issue of how we treat people was a recurring theme in the interviews concerning child-led research. Here, the treatment of children as competent to lead research, rather than just provide data, was central for Kellett and other users of her approach

Similarly influenced by feminist critiques of the researcher-led agenda and approach, and like Kellett and Kozinets, Gauntlett talked of being partly motivated by a desire to avoid research ‘where you sort of go in, get data, and leave’. He was concerned with creating a ‘fair kind of relationship’ and an interesting and meaningful experience for participants. Following his influence, interviewees using his approach did so in part because it makes research fun (Rev4) and because people enjoy the experience and become relaxed (ECR2). This was not altogether altruistic, however, as enabling participants to ‘get caught up in this more bodily task’ of making in which ‘they leave behind some inhibitions’ (Rev4) was ultimately about getting better data from them. For one interviewee, such a stance would be right as, she argued, choice of research methods should not just be because it is the best method for addressing the research question (Exp4). While giving people time to think was part of the rationale and appeal of creative methods, in terms of both research ethics and research quality, this facility was recognised as not peculiar to these kinds of methods (Disc3). Indeed, one interviewee cautioned about the need to use methods carefully and reflexively and to take care in making claims about them. The implication here is that the relationship between being innovative and being ethical extends to incorporate the size and nature of the claims made in the process. Intrinsic to some of the reflections was a concern with whether innovators should be making claims to one method being better than others rather than being just another tool in the tool box to choose from. From this perspective, trying to distinguish innovative methods from traditional ones may be an oversimplification, and it follows perhaps that trying to discern a relationship with research ethics is problematic. It also, however, leads to a second theme emerging from the data – that of innovations reflecting a desire to act responsibly.

Acting responsibly

It was common for our interviewees to consider whether the actions of themselves and others as researchers were responsible. In thinking about his creative methods, Gauntlett was sensitive to who can analyse whose data, wanting his participants to interpret their own Lego constructions. One adopter of the method sympathised; she wanted to avoid her analytic process leading to her feeling like she was ‘damaging’ or ‘trampling’ over participants (ECR2). Another saw this concern to avoid violating participants’ words as a failure to undertake analysis and to really address the overlaying of the researcher’s interpretation over the model-maker’s interpretation. The argument was continued by the suggestion that it is ‘a bit irresponsible of us to assume somehow that data will speak for itself’ and that ‘the subjects of our research will somehow speak for themselves’, as there is ‘always going to be a process of, of selection and interpretation going on, even if that’s not always explicit’ (Disc3).

In some ways, netnography was perceived as a response to the potential offered by technology to behave badly – to treat information available online as data without having to negotiate terms, thereby making obtaining consent a nicety rather than a necessity. Kozinets’ approach is explicitly about an ethical response to a challenge arising from the fact that other, similar approaches do not address the ethical and procedural issues he seeks to solve. He explained that in his early work, some of ‘the biggest questions’ that arose were around ‘Is this ethical?’. Developing rules for working were important to his developing approach, though he recognised that as more people adopt and adapt ethnography, such rules become ‘a little bit looser’ and that ‘if that’s acceptable to the editors and to the institutional review boards’ then he could ‘live with it’. Kellett covers similar ground, but with a more hard-line stance as she stressed the importance to her of others acting responsibly in their take-up of the idea of child-led research; she spoke of wanting them to do this in a responsible way in which they provide more opportunities for children and young people to train as researchers, while at the same time retaining the purity of the ideas, not diluting them.

Kozinets spoke of an agenda to bring to the method and to the marketing/business world ‘an anthropologist’s deep respect for the diversity of human gatherings’, ‘bringing a more respectful view of consumers to those who are ostensibly there to serve them’. For him, being rigorous is acting responsibly:

Having a deep experience of … participating in that community as a member does, rather than, you know, artificially sort of just downloading and coding, or downloading and throwing it into some programming that’s gonna give you some, you know, word map.

For one interviewee (Disc1), it was the temptation to not act responsibly that was largely countered by Kozinets in his approach. Disc3 was aware that in online worlds ‘you don’t have to represent who you are’ or ‘give much information about yourself’ but argued that netnography addresses such ethical tensions. Similarly, another commentator observed that ‘all of the online research methods have hit upon common sets of problems, which they have to solve … or learn to live with’, but innovators like Kozinets have done ‘some of the thinking’ about the ‘huge ethical issues’ (Rev2). Another interviewee explained that ‘there are different schools of thought on whether it is ethical as a researcher to kind of lurk within these communities, or whether or not you need to disclose your identity’ (Int1); the implication of this was that if netnography becomes more popular, the ethical debate will need further attention.

For the interviewees who expressed reservations about the innovators’ responsible actions, one concern was whether the innovators engaged with critiques of their work (though Kozinets regarded this as essential to bolstering a method and achieving academic legitimacy). For some interviewees, this was seen as a responsible thing to do but not always done (Rev4). This again connects with overclaiming in which ‘maybe the enthusiasm is going a bit too far’ (Exp4), bringing the danger of proliferation of a technique without people thinking through the underlying epistemology and ontology.

Democratisation – diverse experts

Common across the cases, and reflecting broader trends in the social sciences, was a willingness to see participants (or child researchers) as having an alternative, legitimate expertise to that of academic researchers. For Kozinets, this was why immersion in online discussions of their experiences and lives is important. For Gauntlett, it was why the researcher should not interpret the Lego models of their identities that participants make, but rather leave that job to the participants who, he said, were ‘the expert on their own lives, or at least, on the thing that they’ve made … with the intention of telling you something about their own lives’. For one adopter of the method, this use of Lego was important for breaking down the traditional researcher–participant ‘power dimension’ (ECR2). For Kellett, of course, children’s expertise is central to her drive for child-led research. She was adamant that children ‘create knowledge that we [adults] couldn’t necessarily create. And they can analyse in a way that, we don’t analyse, because … sometimes they see things that we don’t’.

This valuing of expertise beyond that of professional researchers can be regarded as an ethical stance, though it is rarely explicitly communicated as such. It relates to Lincoln’s (2005) ‘democratic and pluralistic ethics of qualitative practices’ (p. 165), and it encompasses the recognition that people are experts by experience. For Kellett, children have ‘a better understanding of what contemporary childhoods are’ by dint of living them; this makes them distinct and leads to a privileging of their insider perspectives. Her drive to train children to conduct their own research, she explained, reflects not just their perceived competence to do so but her desire to get ‘back to the raw purity of just what the kids see through their eyes’ without adult mediation. There was urgency about children setting the research agenda, because what they think is important might not be the same as that which adults prioritise for research. This is where Kellett was seen by those we interviewed as pushing beyond just valuing children’s expertise into giving them real power ‘to produce knowledge about children’ (Int2). Interviewees (Int2, Disc2) saw how this connects the innovation with feminist research and with emancipatory research in the disability field, where it would no longer be seen as innovative.

It was clear to the interviewees that it is the extent to which Kellett pushes the commitment ‘to working with children in a much more engaged, equal way’ (Disc2) and the drive to put children at the centre of research that is innovative, rather than the idea itself. There was real timeliness in the methodological approach as it links so closely to the ‘participation agenda nationally … and the movement towards children’s rights internationally’, reflecting ‘years of people talking about the child as agent’ within the new sociology of childhood (Disc2). There were also mentions of the echoes of the trend towards involving lay researchers, and this interviewee recognised the tension that once we teach children research skills, their positioning as lay people changes as we encourage them ‘to look at the world through our lenses’. Another (ECR1), though, found the professional angle to involve children appealing.

While supporting the essential ethics of valuing children’s active engagement in research, one interviewee questioned whether children should be leading research regardless of the research question. The perceived danger was of becoming ‘caught up in a zeitgeist of thinking this is what we should be doing’ (Disc2). Another, applauded the ethical dimension of children’s research engagement, seeing this as an ‘extremely good example of taking children’s voice very seriously’, wholly appropriate in the context of Article 12 of the UN Convention on the Rights of the Child (Exp3). Seeing himself as the leading edge of involving children as active participants in research for ethical reasons, he acknowledged how Kellett has pushed further. He also saw the problematic nature of the whole issue of ‘what is that insider knowing that children have, and how it relates to various modes of adult knowing’. The idea that children have ‘privileged’, ‘utterly unique knowledge’ was questioned from a social constructionist angle in which one needs to understand that ‘children’s ways of knowing other children, are themselves informed by all the other discourses that are around’, ‘not devoid of influence from the adult world’ (Exp3). It was put to us that children do not have the only view on childhood and that their voice is, in any case, mediated by research techniques. Even among those highly sympathetic to the ethics of this innovation, the realist epistemology and the boundaries created (between children and adults) were troubling enough for them to resist privileging children’s standpoint knowledge to the extent that Kellett does in her work.

Democratisation – research methods to reach everyone

Linked to, but distinct from, the idea of democratisation of knowledge production through valuing others as experts is the idea, also large in these data, of democratisation of research through methods that reach everyone. Gauntlett was explicit about this; he sees real value in Lego as levelling the field of research methods – a material everyone can use regardless of professional or other status. He spoke of his interest in the potential of Lego to draw in people who may be inhibited by notions of their inability to articulate or draw because the medium is less limiting and ‘people can formulate what they really think’ and what they want to say about an issue. They can also ‘reflect on themselves more clearly, by having gone through that process’; that ‘everyone’ can work in Lego was central to Gauntlett’s thinking and an ethical driver behind the method.

The reality of the ethic of everyone was not shared by all the interviewees reflecting on Gauntlett’s creative methods, however. While one experienced methodologist and supporter of creative methods similarly saw Lego as ‘a neutral medium’ (Exp5), this notion was also critiqued as not necessarily suited to ‘people who don’t feel artistic’ (Exp4). Nonetheless, an attraction for those using the approach was its more participatory approach and potential to reach groups less suited to a straightforward interview. One user of the method talked of ‘getting at those more subconscious or unconscious impressions’ (Rev4), though he saw Lego as one of the many options for this. Another talked of ‘the combination of giving people time, creating an environment of play, and then having the kinaesthetic experience of actually moving the fingers and thinking’ as giving people ‘a kind of permission to talk and express their ideas’ (ECR2). This, in part, is where the claims are seen by others as too bold, when Gauntlett and others make judgements and say that compared to other methods these represent a ‘more rich and meaningful way of asking people to express things’ (Gauntlett), a way of ‘unlocking’ (ECR2).

For those who do use the method of making models with Lego for its power in helping people to express themselves, an ethical dimension was, however, raised by this very power. Gauntlett himself acknowledged that it can be ‘emotionally revealing … getting a bit too intimate or personal’ at times as people open up and make an ‘emotional or meaningful investment in the metaphor’ created. There are strong criticisms though that using Lego in the way described does not ‘unlock’ ‘essential knowledge’, rather, the models constructed are a reflection of a specific sociocultural context (Exp4, Disc3). ¹

The idea of reaching everyone is important in Kellett’s work too. Users of her ideas spoke of how she has wanted to extend, to as many children as possible, the opportunity to do research, reaching down into younger groups than previously involved (Exp2). For advocates of child-led research, there was an agenda about getting marginalised groups heard (Int2), but there was also some recognition that even within child-led research, ‘only certain people get to take part in participation agendas’ with the risk of kinds of exclusivity (Disc2).

Empowerment

The theme of research as empowerment was strong in the creative methods and child-led research cases. With regard to the former, one experienced commentator identified that the notion of creative methods rests ‘on a certain notion of research as a form of empowerment’ (Disc3) with researchers empowering participants by giving them means to express their voices. This interviewee sympathised with the political intention but noted that the notion of reaching the participant’s true identity is flawed (‘Identity is not something we have … identity is something we do’, Disc3). Moreover, he claimed that it is still the case that Gauntlett and other researchers retain their power and define the parameters. Gauntlett himself denied making claims about empowerment but identified a desire to aid participants’ voices by offering a creative process in which they can edit along the way of making their product and using it to explain and reflect. Championing the approach though, Exp5 argued that using Lego, as Gauntlett advocates, ‘empowers’ participants by giving them some control over the interview process.

With regard to developing child-led research, Kellett was clear about her empowerment goals. She talked about promoting the method and the ‘empowerment principles’ arguing that ‘I guess my driver, what gets me out of bed in the morning, is empowering children’. She reflected on seeking to ‘facilitate access to really prestigious platforms to get them [children] noticed’ and to give voice to children and their research. For Kellett, this was a matter of entitlement – children have a ‘right’ to a ‘valid research voice’ – though she perceived fear of children’s empowerment and foresaw a backlash ‘if children become too agentic’. For interviewees with great respect for Kellett’s work, this empowerment discourse was nonetheless problematic. One interviewee commented, ‘I’m sure it attempts to empower children’ but qualified this by noting that the idea of some people empowering others raises all kinds of questions as ‘empowerment itself is such a vexed issue’ (Int2). Kellett’s efforts were seen to have gone into praxis rather than theorisation, leaving more work to be done on theorising empowerment in children’s research. Notwithstanding this, and some questions regarding how, paradoxically, the research methods in which Kellett trains child researchers are very traditional, Exp3 summed up that ‘in terms of empowering children’ her work was ‘radical’.

Relationship with the academy and formal ethics review

We began our discussion of the interaction between methodological innovation and ethical research practice by highlighting the role of institutionalised research ethics regulation. Here, we return to the relationship between innovation and the academy by drawing on the data from this study on this theme. We have argued that innovation may be influenced by drives for democratisation but some of our interviewees questioned whether there was any broadening in access to the privileges of the academy. Int2 noted that ‘Academia is a pretty institutionalised place to bring children into’. Indeed the whole development of child-led research brings into focus the role of the academy in gatekeeping practices and standards, and Int2 saw the ‘benchmarking processes’ in universities as marginalising innovations like co-authoring with children. This is awkward, of course, for as Disc2 pointed out, ‘it would have been very cynical of the academy to be pursuing this child-as-active-subject theoretically and conceptually, but continue to hold the research practice to themselves’. Nonetheless, it raises critical questions about ‘who gets to count as a credible researcher’, and it highlights the ‘tension between participation and rigour’ (Disc2) when other kinds of experts enter the academic domain. One measure may be whether children’s research is published, which another supporter of this kind of work had found to be difficult, especially if children are co-authors, when articles are rejected as not ‘sophisticated enough’ (Exp2). Another key measure may be whether the innovations ever attract research funding: it is ‘not enough for the academy to believe that it’s a worthwhile methodology … there has to be a cultural acceptance that lay-, whether children or not, led research is acceptable … and I don’t think we’re anywhere near that’ (Exp3).

Kellett explained that she covers ethics in the first day of her 3-day training for child researchers and aims at the ‘same level of ethics as adult research’. Managing ethics approvals is one benchmarking role the academy retains, and Kellett was greatly respected by one interviewee (Int2) for battling through and getting ethics approval for child researchers operating through the Children’s Research Centre. She noted how in other countries, including her own, Australia, the formal ethics processes were so institutionalised that this would not be possible. Usually concerned with protecting children from researchers, it is an unusual situation for ethics committees to review research by children, turning some of the surveillance-protection discourses and practices on their head.

Kozinets reported finding that Institutional Review Boards, conscious of the ‘potential for harm’ in online research ‘had no idea how to handle this at the beginning’. One interviewee noted that what Kozinets did not do, ‘which North American Social Scientists are very prone to do, is to get hung up on the whole regulatory governance thing’ (Rev1). His response to the ethical issues had been to try to understand them ‘and then doing what you have to do’. It was, Rev1 said, about ‘being ethical first’ with ‘ethical compliance’ contained within this, rather than the regulation hijacking of the ethical discussion. The innovators in these cases seemed to be tackling the ethics issues head-on rather than simply paying lip service to them.

Discussion and conclusion: innovation, ethics and risk

There are limitations to our research, which has not involved particularly innovative methods or observed methodological innovation in situ. It has, however, involved in-depth discussion of processes in their contemporary and historical contexts and, in a way that has not been done before, explored different rationales, constructions and complex stories surrounding innovation and ethics. We have taken research ethics to be much more than the basic anonymity and confidentiality that can sometimes preoccupy ethics review boards’ protectionist discourses and used the three case studies of qualitative researchers making methodological innovations to explore the relationship between innovation and ethics.

Analysing our data, we found that ethics are important to the innovators and to those who respond to, use and regulate their methods. Their innovations are not a response to unethical practice (in the United Kingdom, the ESRC (2005) has acknowledged there is little evidence of such), but they are in some ways a response to the potential for less than ethical practice and response to the drive to be as fair and participatory as possible in our methods. A characteristic in common across the three areas of innovation is that they are operating in what are often perceived to be ethically risky domains (the Internet, children and visual methods).

In pushing at the boundaries of established methodological practice, the innovators are involved in managing (not eliminating) risk related to their own career development and standing (see the characteristics of innovators discussed by Klein (1990) and Hesse-Biber and Leavy (2008)). We acknowledge that the innovators also risked making themselves vulnerable by taking part in this study and our own responsibility to act with integrity in this process. The culture of ethics regulation further adds to the need to manage the riskiness of methodological innovation. Interestingly, Gauntlett, and to a greater extent Kozinets and Kellett, have sought to codify their methods, creating step-by-step guides (Kellett, 2005; Kozinets, 2010). These may not conform to notions of innovators being creative but instead reflect the cultures in which these academics are operating in which research ethics, for good or ill, have been bureaucratised. The ESRC Framework for Research Ethics (2010) is emphatic: ‘Risks should be minimised’ (p. 3), and this is significant given the powerful status of the ESRC as a funder and influential voice on social science research. The innovators communicate to others that their innovations are contained, not too dangerous and definitely not ethically risky. Making a claim about, or positioning oneself as engaging in, ethical behaviour may balance out the risks associated with any claims to an identity as an innovator. Sikes and Piper (2008, 2010) criticise ethics review committees for positioning researchers as irresponsible; these case studies show the opposite to be the case in demonstrating the researchers’ strong commitment to act responsibly while moving forward methodologically. As we hope we have begun to illuminate in this article, both ethics and innovation are about reflexivity as well as technique, and here, we foster reflection on the relationship between the two. Moreover, we show how the relationship, characterised by an evident tension may also be characterised by a reciprocity, which has hitherto been relatively unexplored.