Digital stories for transition: co-constructing an evidence base in the early years with autistic children, families and practitioners

ABSTRACT Concerns have been raised about the quality of practice-focused research in education generally and in early years education specifically. Chris Pascal and Tony Bertram argue that a shift in worldview is needed to improve the robustness and overall quality of participatory research in the early years and proposed a “praxeological framework” for research comprising praxis, power, values, and methodology. This paper provides an example of how this praxeological framework was applied within an existing research-practice partnership focusing on autism education in the early years. We used a “non-orthodox” Digital Storytelling methodology to co-construct knowledge between researchers, practitioners, children and families about educational transitions. Our co-construction of knowledge involved the embodied knowledge of children and the exemplary (practical) knowledge of families and practitioners, leading to new insights into educational practices. In adopting a knowledge co-creation approach from the start, we established a powerful pathway to impact through which our research is already making a difference to practice. We propose that pathway to impact is an important element that could be made more explicit within a praxeological framing of research.


Introduction
There is longstanding recognition of the value and importance of close-to-practice (CtP) research in education generally (e.g., Rudduck & Hopkins, 1985) and in early years education specifically (e.g., Newman & Woodrow, 2015). Grounded in action research and participatory research traditions, there is an understanding that education as an academic discipline is inherently and inevitably intertwined with practice in many and varied theoretical and methodological ways (Biesta, 2007). Knowledge is generally accepted as situated; generated from specific contexts imbued with cultural meanings (Daniels, 2001). Accordingly, learning processes and experiences are most authentically examined in their context of implementation which necessarily requires the consideration, involvement and engagement of practitioners. Not only is practitioner engagement in research considered a vital ingredient in the development of a "selfimproving" teacher workforce (BERA-RSA, 2014, p. 5), but it is also required for the generation of new knowledge in education research that appropriately recognises, respects, and reflects the individual and local contexts of inquiry (Cochran-Smith & Lytle, 2006;Thomas, 2012), as well as the expertise or craft knowledge of practitioners (Hammersley, 2005;Nind, 2006). In short, "Education research becomes valuable only when it takes account of the reality of the educational endeavor [sic]" (Thomas, 2012, p. 26).
However, concerns have been raised about the quality of CtP, or practice-focused, research; in particular in the most recent evaluation of research quality in the UK: the Research Excellence Framework (REF) 2014 (www.ref.ac.uk/2014). Therein, some CtP research was noted to have " . . . lacked originality, significance and rigour" (Research Excellence Framework, 2015, p. 105). In response, a rapid evidence assessment of CtP research involving literature review and interviews with seven participants (Wyse et al., 2018, Summary pp.1-2) defined high quality CtP research as: . . . research that focusses on aspects defined by practitioners as relevant to their practice, and often involves collaborative work between practitioners and researchers . . . [and which] . . . requires the robust use of research design, theory and methods to address clearly defined research questions, through an iterative process of research and application that includes reflections on practice, research and context. While this definition is helpful in reminding readers that the fundamental qualities of good research should apply to CtP research too, it is agnostic about taking a position on the nature of the research-practice relationships envisaged in its "collaborative work". This matters since the strongest, world-leading, practice-focused research from REF 2014 was found to feature " . . . co-production or close collaboration between learners, teachers and researchers" (Research Excellence Framework, 2015, p. 109). Such co-production or collaboration usually requires taking more "synergistic" (Leibowitz et al., 2014(Leibowitz et al., , p. 1258 approaches to research using inclusive or participatory methodologies which may involve practitioners in different ways (e.g., see Parsons et al., 2015;Seale et al., 2014). These approaches move beyond conceptualisations of knowledge generation that rely on transfer or exchange. In other words, conceptualisations that assume that knowledge flows either from "academic research" to inform practice or supports mutual dialogue between practice and research such that academic research can inform practice (rather than the other way around; see also ESRC, n.d.). This raises important questions about what good collaboration or co-production could look like in high quality, practice-focused research.
Moreover, no examples from early years' education were included in Wyse et al.'s (2018) analysis of the literature suggesting that none met their quality criteria 1 ; and none of their interviewees were described explicitly as having early years expertise, resulting in early childhood research being particularly underrepresented in the report. We therefore wanted to explore an example of collaboration in early years research to illustrate what is possible when research and practice work in co-construction on questions that matter to early years practice. Specifically, we apply a "praxeological framework" from Pascal and Bertram (2012) which we argue is a helpful way of considering the collaborative nature of practice-focused research since it includes the values that underpin the research and, accordingly, takes an explicit position on the nature of research-practice relationships.

Conceptualising collaboration: knowledge co-construction
First, we need to be clear about our own use of terminology. The Research Excellence Framework (2015) report used the term co-production to describe this key feature of high quality CtP research but did not define it further. Armstrong et al. (2019Armstrong et al. ( , pp. 1299Armstrong et al. ( -1300 defined co-produced research as " . . . creating a shared space where the right people are involved, everyone can be equal and equally involved, and we can be adaptable to new ways of thinking and doing research". Similarly, Nind et al. (2017) defined co-production whereby " . . . researcher and research participant can share in the design and conduct of the research and production of knowledge" (p. 392). Thus, the focus of co-production tends to be with how research is conducted, and we are interested in this too. As noted earlier, we argue that knowledge creation is also important to understanding the value of collaborative research; in other words, what knowledge is created as well as how it is produced. McFadyen and Cannella (2004, p. 737) define knowledge creation as: . . . dependent on the combination and sharing of tacit knowledge. For example, in knowledge transfer people often seek answers to specific questions, or seek out specifiable information that they are aware they need. In knowledge creation, information exchange is frequently emergent, in that partners to the exchange are often unable to articulate, a priori, the specific knowledge that they need. This kind of exchange requires very direct interaction as the parties grapple with research puzzles. In effect, new knowledge emerges through the direct interactions of research partners.
In recognising the intertwined nature of knowledge creation and research co-production we, therefore, prefer to use the term co-construction in our own work. This encapsulates our approach based on the praxeological principles discussed further below and is also in alignment with previous work Parsons et al., 2015). For us, coconstruction encompasses new knowledge creation (the what) through the shared endeavours of research and practice working together equally (the how).

Co-construction as praxeology in the early years
There is a rich and productive record of practice-focused research in early childhood education (e.g., Lyndon et al., 2019;Newman & Leggett, 2019). There is also no shortage of critical reflection on the field, and its nature of evidence (e.g., Anning et al., 2004;Newman & Woodrow, 2015). Pascal and Bertram (2012) reflected on their own contributions throughout over 20 years of early childhood research and suggested that there is a " . . . need to respond to a continued and sometimes reasonable professional critique of the robustness of our participatory methods" (p. 479). Pascal and Bertram (2012) proposed that a shift in perspective or worldview was needed to conduct early childhood research that is " . . . more democratic, participatory, empowering and . . . deeply ethical and political in its orientation" (p. 479). Drawing strongly on Freire (1970), Pascal and Bertram (2012) were explicit about their social justice positionality viz. that education research should seek to be transformative, and to impact on practice in ways that incorporate and value the perspectives of all stakeholders including practitioners, children and families. Such approaches are inevitably, therefore, strongly collaborative and recognise that knowledge is situated, subjective and co-constructed and that research should be " . . . practised by those who are committed and close to the real world of children and families" (p. 484). They called this adopting a " . . . 'praxeological' worldview in modern early childhood" (p. 480) which involves more than simply describing or doing activities and, instead, focusing on purposeful action in a way that requires: . . . reflection (phronesis) and action (praxis) done in conjunction with others . . . [and] immersed within a more astute awareness about power (politics) and a sharpened focus on values (ethics) in all of our thinking and actions. (Pascal & Bertram, 2012, p. 480) This stance aligns strongly with related research approaches such as inclusive research (Nind, 2017), and the "new science of education" proposed by Thomas (2012, p. 26), both of which align with the foundational principles of action research (Rudduck & Hopkins, 1985) and participatory research involving practitioners and researchers more generally (e.g., Christie & Menter, 2009;Leibowitz et al., 2014). Inclusive research as defined and discussed by Nind (2017) is concerned with a greater democratisation of the research process, particularly in relation to including the situated knowledge and lived experiences of people with learning disabilities, whose voices and experiences have traditionally been marginalised. Similarly, Thomas (2012, p. 28) critiqued the "unyieldingly monistic view" that evidence in education should conform to ways of gathering evidence and judging its quality based largely on ideas that come from different scientific domains, especially medicine. Acknowledging the importance of the local and the individual in knowledge generation that matters to educational practice, Thomas (2012) argued that " . . . our landscape of inquiry exists not at the level of those big 'what works' questions but at the level of personalized [sic] questions posed locally" (p. 41).
Thomas (2012, p. 43) discussed the vital role of "exemplary knowledge" in the generation of practice-focused evidence, and theory building around such evidence, which is based on sharing experiences with others to reflect on, and learn from, different perspectives and gain insights into practice. This is a process of knowledge-making that Cochran-Smith and Lytle (1999) describe as " . . . inevitably a process of theorizing" (p. 272). Such "exemplary knowledge" seems to encapsulate Pascal and Bertram's (2012) "reflection . . . and action . . . done in conjunction with others" as quoted above. In other words, individual reflection on action is not enough for knowledge generation within the praxeological framework but rather needs to be done with others to build exemplary knowledge. It is knowledge co-construction, therefore, that is key (Freire, 1970).
Our aim is not to make fine cuts between these ways of framing education research that are grounded in practice and stakeholder concerns. They share a fundamental ethos in relation to knowledge creation and how, where, and with whom such knowledge creation should be achieved. Instead, our aim is to highlight and draw on Pascal and Bertram's (2012) praxeological framework through applying it to our own research with autistic children in the early years to illustrate what is possible when research and practice work in co-construction on questions that matter to early years practice. As Pascal and Bertram (2012, p. 480) suggested " . . . we . . . invite you to contribute to the dialogue", and it is in that spirit that we are presenting this work. Specifically, Pascal and Bertram's (2012) framework focuses on: reflection and action; power redistribution (politics); values (ethics); and methodologies. We also propose an additional element to the framework focusing on impact. An overview of this framework and its application to our own work is shown in Figure 1.

The need for a praxeological turn in autism education research in the early years
We propose that there is a context of multiple marginalisation of practice-focused evidence in autism education research in the early years that is very much in need of a shift in worldview along praxeological lines. Space precludes a full rehearsal of the argument, but readers are referred to Guldberg (2017) and Guldberg et al. (2017) for detailed discussion. In essence, autism education research primarily tends to be steered by the big "what works" questions of which Thomas (2012) and others (e.g., Biesta et al., 2014;Malone & Hogan, 2020) are so critical, along with concomitant expectations about the nature of research designs and high quality evidence. Autism education research is dominated by biomedical approaches that are often unconnected to education practice directly through either not being conducted in educational contexts (Parsons & Kasari, 2013) or with any involvement of educational practitioners (Guldberg, 2017). Indeed, Conn (2015) argued that much autism education research would benefit from "context-thinking" rather than " . . . assumptions . . .
[being] made about children's social worlds" (pp. 66-67). Research funding trends internationally, and especially in the UK, show that autism research focuses mostly on biology and the brain, and intervention and treatment, rather than educational practice or provision (Office of Autism Research Coordination, National Institute of Mental Health, Autistica, Canadian Institutes of Health Research and Macquarie University, 2019). Within this context, the voices of autistic 2 people, and especially young autistic children, are recognised as marginalised (Ridout, 2017). While there are recent efforts aimed at addressing such marginalisation (e.g., Chown et al., 2017), the autism field lags a long way behind that of disability research more widely which has been grappling with fundamental issues of representation, engagement, and knowledge generation for over 40 years (Oliver, 2013).
Consequently, within an already diminished space for representation in the evidence base, the voices of autistic children are especially absent (Cascio et al., 2020;Ellis, 2017). We argue this is due to prejudices which assume that autistic children find it difficult to express their views because (a) they are children, and (b) they have a (social and communication) disability (Lundy's 2007 "double denial" of voice, p. 935), and that (c) this combination of being both a child and having a social communication disability positions them as being especially "hard to reach", even amongst other disabled children (Franklin & Sloper, 2009, p. 4). Indeed, LeFrancois and Coppock (2014, p. 166) argued that the biomedical dominance of research about children diagnosed with psychiatric disorders (including autism) denies them their rights as "knowers" in their own lives. While autistic children's voices are beginning to be gathered and contribute to our understanding of their experiences, the voices represented are frequently from older, and more articulate children and young people (Fayette & Bond, 2018). Combine all these influences with the low status typically accorded to early childhood practitioners, and the limited opportunities for professional learning in the early years' workforce (Newman & Leggett, 2019), and there is an undeniable confluence of multiple strands of marginalisation in early years autism education research that needs to be challenged. Praxeological Framework for high quality close-to-practice research

The voices of young autistic children preparing for transition: applying and extending the praxeological framework
In seeking to enact an opportunity for practice-focused, theoretical construction in the early years, our own work was situated within a research-practice partnership established between academics in Education and Psychology at the University of Southampton, and local education providers (nurseries, schools, and colleges) working collaboratively to coconstruct a more meaningful, and locally relevant evidence base for autism education. The Autism Community Research Network @ Southampton (ACoRNS) was established in 2017 and its partners span statutory educational provision across mainstream and specialist contexts. The importance of the local context and the situated nature of knowledge was recognised from the start as vital if we wanted to influence and improve practice. Together, we pose research questions that come directly from practice and seek to answer them through mobilising the research expertise and resources of the University in conjunction with practice-based expertise and experience (see Parsons & Kovshoff, 2019, for more details). Together, we establish shared priorities and work out how best to address them via research within our local contexts of practice. Crucially, children's voices and perspectives are at the heart of our endeavours. In short, ACoRNS provides an active space in which we (researchers and practitioners, alongside children and parents) can "think together" (Conn, 2015, p. 61) about the issues that matter to us and the children and families we support. Under our ACoRNS umbrella, we draw upon our Froebel-Trust funded research to illustrate a praxeological approach in action with Aviary Nursery, who were one of the founding partners of ACoRNS. The project focused on accessing the voices, views and perspectives of young autistic children as they prepared to transition to their first school after nursery, and was aimed at addressing the longstanding issues, summarised above, relating to the underrepresentation of autistic children's views in education and an underestimation of their capabilities. Not only is this approach important in terms of according autistic children the same rights and respect as non-autistic children to be "active co-constructors of meaning in their own lives" (Pascal & Bertram, 2009, p. 255), but seeking children's views directly about their educational provision is mandated in England through the Special Educational Needs and Disability (SEND) Code of Practice (Department for Education/Department of Health, 2015). Thus, there are moral (social justice) and legal obligations for developing methodologies that position young autistic children as knowers in their own lives (see Parsons et al., 2020, for a project overview).

Power redistribution
We begin with power redistribution since without this it would be difficult, if not impossible, to reflect and act together in ways that genuinely support the coconstruction of knowledge. Through the ACoRNS partnership being positioned as coconstructed from the start, there was an explicit expectation and commitment to challenging the dominant, traditional orthodoxies of ways of doing research, which includes shifting power relations towards a more shared endeavour (Freire, 1970;Tomlinson & De Ruysscher, 2020). For us, there were vital catalysts, or pivotal moments, in the formation and development of ACoRNS that were crucial for shaping our own power relations in important ways and directly influenced our work in the early years. The first pivotal moment was when Aviary nursery colleagues contacted us at the University with pressing questions for research coming from their own practice: "How do we access the voices of young autistic children who may not communicate in typical ways?"; and "How do we make sense of their perspectives and experiences?" There was a strong sense of the need to think together about how we could answer these important questions. The initiation of questions from Aviary coincided with conversations with another existing collaborator (New Forest School), who had also approached us at the University with interests in working together. Building on prior experiences within the team in connecting research and practice agendas , this led to a small-scale funding application to the University's Public Engagement with Research unit (PERu) to establish a more formalised partnership between the University and local educators who shared a commitment to addressing research questions that come from practice. This was our second pivotal moment. Fortunately, we were successful with this application and set about agreeing our research priorities together which settled on the "transitions and trajectories" of autistic children in education.
As we argue elsewhere (Parsons & Kovshoff, 2019), a critical element in the success of research-practice partnerships is the initiation of interest and ideas that come from practice rather than from a research agenda which may then feel imposed on educators. However, this does not mean that everyone contributes in the same ways or that power is shared tidily across all aspects of our activities; indeed, an imposition of the latter would have been detrimental to our own progress and, therefore, opportunities for learning. The tensions around the redistribution of power in participatory research are well rehearsed. For example, Pascal and Bertram (2012, p. 488) discussed the vital role of "transformational leadership" within the navigation of roles. The tension exists because on the one hand it takes vision, skills and leadership to move ideas forward and to make things happen and someone must lead these endeavours. On the other hand, there is inevitably a risk of exclusion of others through the dominant perspective and actions of leadership. Thus, transformational leadership requires a "new way of being" (Pascal & Bertram, 2012, p. 488) that is developed over time through an iterative to and fro of thinking, sharing and doing. As co-author Kathryn Ivil (the Manager at Aviary Nursery and a founding member of ACoRNS) makes clear, this has worked well so far due to our focus on a shared interest as well as the sharing of roles: From a personal perspective, the opportunity to share how important, but currently overlooked, Early Years practice is and how skilled and knowledgeable most early years' teams are, has been a significant part of this venture. The ACoRNS team have valued my views, knowledge and input which does not always happen as an early years' practitioner . . . This has been beneficial personally but has also reached other members of The Aviary team and made them feel valued as well. The ACoRNS partnership has been successful due to the mutual respect for each person's skills, knowledge, experience and personality. This transformational model has transpired due to the personalities of the team and a shared desire to research significant areas which are relevant now and provide the very best care and education to some of our communities most vulnerable children.

Reflection and action in conjunction with others
The initial questions that came from Aviary were vital in orienting us towards research priorities coming from practice, and in trying to find answers there was an inevitable sharing of knowledge from our different domains of expertise. The idea of co-applying for funding, and writing and submitting the grant application, was led by the University researchers, building on experiences and ideas from some of our earlier work . The inclusive "pedagogic culture" of the nursery (Arnott & Duncan, 2019, p. 1), its excellent relationship with families, and its strong commitment to developing and reflecting on practice, formed the vital context for thinking that a joint project of the kind we envisaged could even be possible. The idea of the Digital Stories as a methodology (more about this below) came from academic colleagues, but the problem-solving with Aviary colleagues around the practicalities of its development and implementation in situ was essential for supporting meaningful and authentic engagement of children, families, and practitioners. We also applied, and pitched, for the project funding together, thereby extending our thinking together through the doing and experiencing together; a good example of Pascal and Bertram's (2012) reflection and action done in conjunction with others. We have also tried to ensure, as far as practically possible, that our dissemination and public engagement work is done together (including the co-authoring of this, and other, papers). Through reflection on this context and process, Kathryn Ivil recognises the value the ACoRNS partnership brings for learning and professional development: This relationship with ACoRNS has given the Aviary Nursery opportunities to be at the heart of relevant research linked to the development issues of the children and families. We have been able to identify areas we would like to know more about, expand or do differently, and proceed in a professional way with the research experience and knowledge of the ACoRNS team. Personally, and professionally, the opportunity to jointly deliver workshops on the work of ACoRNS and Digital Stories, and to apply for funding, has increased my confidence and given meaning to the research and work we do in the early years sector. It has acknowledged and validated that we are important practitioners in our own right and do not need a "teacher" to tell us what to do, we can find out and complete research with the right partners. Pascal and Bertram (2012) are clear about the social justice positioning of taking a praxeological approach; there needs to be clear intent to make a difference for children and families (also Newman & Leggett, 2019). Similarly, Tomlinson and De Ruysscher (2020) argued that in taking a dialogical, co-constructed approach to research " . . . ethics come before methodology . . . rooted in . . . the 'meeting of minds'" (p. 11). In other words, it is not an a priori methodological specification from researchers that drives the research but rather the questions, or lived experiences, that come from practice around which cocreative discussions about methodology subsequently take place. Our own social justice drivers came from a range of influences but converged around the desire to "do better" for autistic children and families. From Aviary's perspective, "doing better" was based on genuine questions from practice around meeting the challenge of the SEND Code of Practice for hearing children's voices about educational provision. It was also driven by a desire to not only be involved, but to influence and contribute in ways that moved beyond simply receiving information and being expected to act upon it. For researchers at the University, "doing better" was based within longstanding interests in, and pursuit of, methods for authentically accessing the views of autistic children and their families as well as developing more participatory ways of conducting research with children, families, and practitioners. There was also strong critical engagement with core concepts around inclusion (educationally and methodologically) and how these are instantiated in practice. Thus, there was a natural alignment in worldviews, without which the ACoRNS partnership would never have got off the ground.

Values and ethics
In being funded by the Froebel Trust we explicitly acknowledge the central role that the Froebelian principles of early childhood occupy in our thinking. These principles promote inter alia: the integrity of childhood in its own right; the holistic nature of the development of every child; the uniqueness of every child's capacity and potential; and the vital role of play and creativity in children's development (Froebel Trust, n.d.). Inevitably, this is a political stance since these principles provide a direct challenge to the impact of the idea of children's "school readiness" that is ingrained within the Early Years Foundation Stage (EYFS) statutory framework in England due to its focus on accountability, academic skills, and standardisation (Hoskins & Smedley, 2019). These principles also challenge the deficit drenched literature on early autistic childhoods noted earlier, by reclaiming a holistic and unique childhood space for autistic children as active agents and knowers in their own lives, including their rights to play and to be creative.
As a team we therefore strongly advocate for a more ethical understanding of childhood, and early childhood education, which draws on a more holistic understanding of the child through prioritising child-led play, exploration, and expression, and argue that this focus should apply equally to autistic children (cf. LeFrancois & Coppock, 2014). It was, therefore, incumbent on us as a team to find ways to get closer to autistic children's experiences and worldviews to enable a fuller understanding of who they are. This is where methodology comes in as a crucial means for intertwining (political) principles with practice.

"Non-orthodox" methodology
As central stakeholders in education, children's views are vital for understanding the kinds of environments that are needed for supporting their positive engagement and experiences such that they can contribute meaningfully to educational decision-making (Murray, 2019;Zilli et al., 2020). As much as possible, therefore, children's views are sought within ACoRNS projects alongside parents' and teachers'. This means being creative about how voices are respected and represented, and competencies nurtured and enabled. In line with others (e.g., Conn, 2015;Ellis, 2017;Goodall & MacKenzie, 2019) we recognise that autistic children will express themselves in different ways and so it is crucial that the onus lies on us being creative and open with the research tools used to enable children's expression and contributions (Morris, 2003) rather than labelling children as inherently "hard to reach" (Franklin & Sloper, 2009). As Mortimer (2004) reminds us: " . . . unless children are allowed to make their needs known in whatever way is available to them, we cannot possibly 'listen' to those children who have no words" (p. 170). Pascal and Bertram (2012) acknowledged that praxeological research, in seeking to actively promote the views of the traditionally "silenced", requires researchers to incorporate more creative, non-orthodox methods to support " . . . alternative and expressive forms of knowing" (p. 489). They suggest dance, mime, drama, storytelling, singing, photography, and film-making as just some of the methods that can be employed. As noted above, we used a Digital Storytelling methodology in our Froebel Trust funded project, drawing on the work of Lambert (2010Lambert ( , 2013. Crucially, this is a methodology that comes from the ancient oral traditions of storytelling that do not rely on writing and literacy as vehicles for knowledge creation and transmission. Such an approach has been further extended and transformed through the affordances of digital technology to incorporate sounds, pictures and video, thereby moving past purely oral/linguistic modes of expression as well. We have explored and developed this methodology within some of our own previous work focusing on the experiences of, and pedagogies with, autistic children Parsons et al., 2015). Therein, we argue that Digital Storytelling is much more than simply a visual method for stimulating a response; rather it is a means through which tacit knowledge can be made explicit and available for sharing and reflection (cf. Rynes et al., 2001) such that the Digital Stories become "evidential artefacts" in their own right (Parsons et al., 2015, p. 251).
More details about our specific methodology in the Froebel Trust project can be found in Parsons et al. (2020). Five children, all boys and aged 4 years, were the focal participants in the project along with their parents and Aviary staff members. We filmed with them over a period of three months to get to know the children and families closely. Aviary colleagues led and guided the local implementation of the project and liaised with families, and the Digital Stories were co-constructed with families and staff through placing children's experiences at the centre of all our thinking. Overall, what the Digital Story creation enabled us to do was focus on the spaces and opportunities for exploration and interaction that provided meaning for the children, from their perspectives. For example, the stories show how meaningful some solitary activities were for the children through hearing their self-talk as they picked daisies, or played with plastic dinosaurs, in contrast to the usually deficit-focused characterisations about autistic children's solitary play in the wider literature (e.g., Anderson et al., 2004). Consequently, we developed the "I am . . . " Digital Story framework that focuses on children's personalities, preferences, agency and capabilities; it is a framework that aims to get closer to understanding who the children are rather than assessing what they cannot do .
A critical element of the methodology that enabled us to do this was the use of individual Wearcams for each of the children in addition to the more familiar way of placing an observational lens on children's lives and filming them from a distance. The idea of using Wearcams transformed the entire project because it enabled us to be alongside children's experiences and perspectives in a way that would otherwise not have been possible. It was an idea that came directly from one of our nursery colleagues, 3 and was another pivotal moment that was borne out of our shared thinking and doing. Inevitably, the use of the Wearcams opened a range of ethical issues for consideration, not least in terms of countering critiques that we could be placing the children and their peers under surveillance and, therefore, making unwelcome intrusions into their lives. As we discuss in Parsons et al. (2020), the Digital Stories need to be viewed directly for us to convincingly demonstrate that we were on the side of the children and that what was revealed about them is powerful and enlightening rather than intrusive or exploitative. The Stories are in the public domain on our website (autismtransitions.org) and so we would always encourage readers to view them to make up their own minds. We are also not complacent about this; clearly the use of Wearcams needs to be applied with much caution, sensitivity and awareness as well as sustained reflexivity.
There was also much practical negotiation that took place to see whether the Wearcams would be appropriate and acceptable for the children and staff at the nursery. We originally discussed Wearcams that could be worn on a headband but dismissed these as visibly obtrusive and unlikely to succeed, and so explored other options. We found some circular (4.5 cm diameter), very lightweight and inexpensive cameras that could be pinned onto clothing and staff were keen to give them a try. Some of the children did not like having anything attached to their clothing and so staff experimented with attaching the camera to a vest that was worn over children's tops. This worked well and children were happy to wear them. We also made it clear to staff what the purpose of the cameras was (i.e. the cameras were not there to scrutinise them) and communicated with all parents of children at the nursery to explain what was happening. Through trusted practitioners taking the lead on the development of this methodology there was clear communication with all involved and a willingness to try out different practical solutions to make this acceptable for the children. Consequently, the use of the Wearcams became possible and was critical to the learning and impact gained from the project.
In "being with" (Morris, 2003, p. 345) the children via Wearcam footage, we argue that we have incorporated the embodied knowledge of the children within our knowledge cocreation approach. Lawlor and Solomon (2017) described embodied knowledge as "a type of body reading that exceeds cognitive and linguistic modes" (p. 238). In a wider context where children's positionality as knowers in their own lives is imbued with so much doubt and deficit (LeFrancois & Coppock, 2014), we therefore must find ways of sharing knowledge that are not inevitably tied to the verbal or written means of expressing it (see Gabriel, 2020, for further discussion). Thus, by placing autistic children's worldviews at the heart of our endeavours, we were challenged to find creative means to enable their expression and voices in "non-orthodox" ways. In so doing, we suggest that our knowledge generation in the early years included exemplary knowledge developed between the researchers and practitioners while also extending our understanding through prioritising children's embodied knowledge. Consequently, children's experiences added substantially to the building of our theoretical constructions of everyday transitions within a nursery setting and the inclusive practices that support children's participation, play, and engagement. For us, this is a good example of how a praxeological framework moves beyond the merely dialogical to push for the inclusion of more embodied methodologies for generating knowledge from those who are traditionally marginalised. Pascal and Bertram (2012) are clear that praxeological research should be transformative for practice such that "Profound change should and does grow from experience to conceptualisation and not the other way around" (p. 484). However, they leave this part of their framework relatively untouched in terms of showing how this might be achieved in practice. We therefore argue for a more explicit characterisation of this transformational expectation within the praxeological framework through including a "co-constructed pathway to impact" as a core element and illustrate here how this has unfolded in our own work.

Making a difference: co-constructing a pathway to impact
In seeking to learn from the initial period of external funding, colleagues at Aviary sought to embed the Digital Story methodology in their everyday practices at the nursery because they could see the difference the project had already made to the awareness and practices of staff. As Kathryn Ivil says: "We did not have to wait to be told what was going to work for us; we could see it happening for ourselves. Digital Stories has become what we do now as one of the ways we show a child's interests and development to other professionals and parents. It is an important piece of evidence which shows research impacting practice immediately". This very much aligns with Sinclair's (2004) observation that for children's participation to be meaningful it needs to be "embedded within organisational cultures and structures for decision-making" (p. 116) rather than seen as isolated or occasional consultations.
Consequently, together we applied for impact-focused funding from the University of Southampton to continue the work, resulting in the piloting and evaluation of embedding Digital Stories in person-centred planning meetings for new children not involved in the original project. This was led by Aviary colleagues, with support from a researcher funded by the grant. Feedback was very encouraging, with parents and teachers recognising the value of the stories, for example: Beautiful. It captured just how he is, which words cannot describe. (Parent) Breath taking! Such a treat to see the world through his eyes and how he goes about his day.
It helped to gain a bigger picture before starting the meeting. It meant that I was able to contribute more to the meeting than if I hadn't seen. (SENCo at receiving primary school) We are continuing to build on this with further funding to extend the methodology to new nursery settings to reach a wider number of children and families. We have also been invited by different Local Authorities to talk about the Digital Stories work with their early years' teams in order to promote the uptake of the "I am . . . " Digital Story framework in practice. Other schools in the ACoRNS partnership are keen to adapt and apply the methodology; for example, colleagues at Hill House School are now working with one of our DEdPsych students to develop Digital Stories with older children with more complex needs, and colleagues at Springwell School have successfully applied to their Local Authority for funding to explore the use of Digital Stories in their setting with children with more complex needs. Under the lockdown for COVID-19, we were also approached by a contact from one Local Authority and asked to develop some web resources to support Digital Story creation since: . . . we are hurtling towards transition time with no idea as to how this is going to look for our numerous very vulnerable children. The glimmer of hope within this is that we need Digital Stories more than ever before as traditional transitions are going to be difficult before the summer.
We will continue to learn from each other as these initiatives progress. Once again, the critical point is that the initiation of interest in taking the ideas forward in practice has come from the local authorities/practitioners/schools themselves. The research has been invited in to practice rather than practice being invited in to the research. We argue that this is a vital element of knowledge co-construction and an indicator that power redistribution is real rather than rhetorical.

Conclusions
In applying a praxeological framework to discussing our co-constructed research within the early years we have aimed to show how greater voice and power can be enabled for those who are usually marginalised: early years' practitioners, young autistic children and their families. These are voices, as we argue earlier, that are typically disregarded within research and practice. We also aimed to illustrate what collaboration can look like in the context of high-quality practice-focused research in the early years. Applying Pascal and Bertram's (2012) praxeological framework enabled us to take a stance on the debate, rather than simply describe what we do. In taking this stance we suggest that high quality close-to-practice research must be concerned with the redistribution of power through negotiation and relationship building, as well as with explicit recognition and instantiation of values that guide actions and thinking together.
This moves us as a field beyond an account of a rigorous and robust methodology (cf. Wyse et al., 2018); while necessary for high quality practice-focused research this is not sufficient in and of itself (cf. Research Excellence Framework, 2015). Rather, this example of praxeological research illustrates a "way of being", and therefore also a "way of doing" research that encompasses and includes the ways of "being" and "knowing" of others, especially those who are regularly undermined and underestimated (Freire, 1970). Thus, we suggest that authentic and valuable knowledge generation and impact in the early years, with a focus on autistic children, can happen when conditions enable the cocreation and sharing of exemplary knowledge (between research and practice) with embodied knowledge (from children, families, and practice). In this sense, there was a "cultivation and contagion of expertise" (Lawlor & Solomon, 2017, p. 234) in our own work that reflected and respected the situated knowledges of all involved. While our work focuses specifically on autism education in the early years, the methodology, framework and methods could be helpful in a range of research contexts in education and beyond.
As Pascal and Bertram (2012) also make clear, such research is inherently risky because it tries to do things differently, through the redistribution and negotiation of power as well as via the use of creative, "non-orthodox" methodologies. Imagine a scenario where researchers, unknown to an early years' context, approached staff to ask whether they could set-up cameras to film in the nursery over a period of months and, moreover, attach individual Wearcams to children in order to capture their movements and interactions? Such research is highly unlikely to have been possible, at least not in an authentic naturalistic way. However, within the context of a trusting partnership where good relationships already existed between research and practice, and between the nursery and the families, we could take the risk together. The new knowledge gained as a function of empowering the voices and perspectives of children and families is powerful and profound, and already making a difference to practice. Indeed, as we have illustrated above, making an impact on practice is a core element of a praxeological approach. Importantly, it is the ethics of social justice that drive the research rather than responding to technical, top-down instruments for accountability such as the REF.
We are not naïve to the challenges with knowledge co-construction however. It is often difficult and time-consuming such that not everyone feels they have been able to continue. In our case, the importance of opting in rather than being invited in to ACoRNS has been played out in practice; it has been much more difficult to maintain power sharing relationships and activities with partners who were invited (and, initially, keen) to join ACoRNS as we were launching the initiative, compared to those who were founding members or who have made contact since, wanting to get involved. We have found that the most successful research-practice relationships are those where the leaders and managers of settings/services are directly involved since they can make decisions and implement initiatives more directly through allocating resources and supporting staff engagement. This is another good example of the principle of power redistribution noted above since these relationships are ones where practice-based leaders have already identified key questions that matter to them and the need to work in partnership with research to find answers, and so there is strong motivation to invest time in the collaboration from the start.
Finally, Newman and Leggett (2019) suggested that "Research that documents their [early childhood practitioners'] achievements can also strengthen the profession" (p. 121). We hope that this paper provides an example of this, and at the very least some food for thought about what can be achieved through working in genuine partnership. We do not claim this is the best or only way to approach the challenge of connecting research and practice more strongly but wanted to share our own learning to illustrate what can be possible. As one of the typically marginalised voices in this space, it is right that Kathryn Ivil has the final words to illustrate what this partnership and project has meant to her: The learning has been reciprocal, with the whole ACoRNS team learning about Early Years, our environment, our current cohort of children and families and some of our challenges. I have learned so much from each person and believe that they have already influenced and expanded my knowledge and understanding of both research and practice in other areas.

Notes
1. Based on the quality criteria of originality, significance and rigour as defined by the REF (see appendix 4 of Wyse et al., 2018, for a summary). 2. In line with the preferences of the UK autism community, the terms "on the autism spectrum" or "autistic person" will be used rather than "person with autism" to represent identity first language; for further discussion see Kenny et al. (2016). 3. We are indebted to Gareth Shaw for this suggestion.