Photo-elicitation and photo-voice: using visual methodological tools to engage with younger children’s voices about inclusion in education

ABSTRACT Whilst the importance of engaging with children’s voices is now more widely recognized, there is still a dearth of representation of younger children’s voices specifically. Visual methodological tools, such as photo-elicitation and photo-voice are used in research; however, there remains a lack of clarity about what they mean, how they are used, and whether they are effective in providing valuable insights into younger children’s everyday worlds. This paper draws on the findings of two research projects utilizing the tools to engage with the voices of 56 children aged four to five years, in the classes of three schools in the UK. It identifies the differences between photo-elicitation and photo-voice and appraises their advantages and limitations when engaging with child voice. It explores whether through careful, critical and conscientious application, these tools might advance understandings of younger children’s perceptions of matters affecting them. The definition of photo-elicitation is the insertion of a photograph by the researcher into a research interview to evoke information, feelings, and memories due to the photograph’s particular form of representation; photo-voice is defined as a technique that enables people to record their own photographs.


Introduction
Despite the growth of recognition of the importance of engaging with child (or student) voice (Ainscow and Messiou 2018;Harris et al. 2017), there is still a scarcity of research with younger children due to the scepticism about their ability to participate meaningfully (MacDonald 2009;MacNaughton, Hughes and Smith 2007). A paradigm shift towards the acceptance that children have views and opinions separate from the adults in their lives (Matthews 2007), repositions them in international policy as key informants and experts on their own lives (MacNaughton, Smith and Davis 2007) and consequently identifies them as the best source of advice for matters affecting them (Osborn and Bromfield 2007). Despite this positive move, there is still some apprehension in eliciting the views of preschool children below the age of eight years (Pound and Lee 2011).
By contrast, there exists a strong tradition of adults conducting research on younger children, mainly within the fields of developmental psychology and education (Tay-Lim and Lim 2013), which often focuses on the identification of developmental deficiencies rather than what the child can meaningfully contribute to research. Mayall (2001) counsels an alternative approach, whereby research is conducted about children, and where researchers invite children to think about the design or involve them in the data collection and analysis to help adults understand their perspectives. Recognition of children as expert informants, who are active in the construction and determination of their own lives (Prout and James 1997), calls for researchers to be authorities in developing and employing appropriate strategies that effectively elicit the insights children can bring to research . Hart's (1992) ladder and Shier's (2001) pathway illustrate the greater or lesser extent to which children can express their views, from verbal or non-verbal means, to active participation in decision-making. Notwithstanding the challenge of engaging with younger children's voices, by eliciting their 'authentic voice' ( Spyrou 2011 ) and achieving deep participation ( Kesby 2007 ), it becomes possible to listen and hear what they say.
Acknowledging Morrow's (1999, 213) suggestion that 'if we are going to listen to children (which is innovative in itself), then we are going to have to be innovative about doing so', more participatory methods need to be considered that are sensitive towards younger children's competencies (Einarsdottir, Dockett, and Perry 2009;MacNaughton, Hughes and Smith 2007;Mukherji and Albon 2010).
Against this backdrop, the paper explores some of the tensions that arise when employing visual methodological tools with younger children. It

Visual methodologies
Child-related research is often based on qualitative interviews, which are regarded as an adequate tool to gain direct access to children's perspectives (Christensen and James 2008; Greig, Taylor and MacKay 2007). Yet interviews may prove challenging for children to verbalise their responses, or find words to describe abstract processes and circumstances.
They may tire quickly or it may prove difficult to keep their interest for longer periods when focusing on non-contextualised topics (Cappello 2005;Gopnik 2009 Many researchers in the education of younger children recommend constructivist strategies as an effective means to elicit views and opinions (Stipek 2004;Nie and Lau 2010).
As Thomson (2008, 1) argues, being young does not mean that children have 'nothing to say', merely that one needs creative ways of listening. By adopting different or multimodal ways, it becomes possible to listen to 'the things that are unsaid' (Thomson 2008, 4) in addition to the spoken words . It is what Rinaldi (2006, 98) describes as 'pedagogical research' -the process of searching for meaning that only the child can offer. She proposes the creation of a culture of research with children, which helps them to reflect and clarify their thinking. Goldman-Segall (2014, 93) discusses 'establishing a rapport' between the child and the researcher, where children view the researcher as an adult, but acknowledge that their role is not to direct the conversation, but to share in the experience; thus, adults and children become collaborators or 'partners in learning'. By starting with the child when seeking to collect data about, rather than from , the child (Coad and Lewis 2004;Waller 2006), adults may share the same story in a context they have jointly constructed.
Engaging children in research, introduces other ways of knowing through using participatory research methods (Cox 2005;Leitch et al. 2007;O'Brien and Moules 2007).
This approach responds to Collier's (2001, 51) concern that research is 'a sea of words and more words' but does not allay his fear that 'visual communications are not taken as serious intellectual products'. With younger children, participatory methodologies foreground the key to unlocking their potential to contribute rich and useful perspectives to inform research into their lives (Tay-Lim and Lim 2013). Placing interviews with children into their everyday activities (Tammivaara and Enright 1986), or encouraging researchers to integrate visual methods of data collection into interviews (Cappello 2005), makes interviews fun and less school-like.
Adding value to existing methods in qualitative research through visual methodologies, introduces another dimension (Balmer, Griffiths and Dunn 2015), which provides valuable insights into the participants' everyday worlds (Barbour 2014). Children's understanding and experience of the world and the way they communicate, is different from that of adults, necessitating an alternative approach (Thomas and O'Kane 1998). Visual methodologies enhance the richness of data by acknowledging and breaking down the disparities in power and status between the researcher and the participant. For younger children, adults can help with developing this relationship, by enhancing rapport building, enabling expression of emotions and tacit knowledge (the unspoken and unexpressed), and encouraging reflection (Pain 2012). Goldman-Segall (2014, 88) refers to a 'culture for shared collaborative "authorship" and distributed co-construction'.
Notwithstanding this potential, participatory approaches may actually advance dominant interests (Hart 1992;Cooke and Kothari 2001) rather than reduce tokenistic consultation . If presented as a set of techniques rather than as an ethical or political commitment, they can actually reproduce unequal power relations (Kothari 2001;Kesby 2007). Photographic images do not empower children on their own; it is the shared construction of knowledge around conversations with the children, based on their photographs, which enables children's meaning to prevail (Cook and Hess 2007).
Thus, data collection that is both child-centred and hands-on, enables children to construct meaning through the research process and share it with researchers (Parker and Neuharth-Pritchett 2006). This process, however, may endorse a view of the child's commentary as an end rather than a combination of both a beginning and end. Cook and Hess (2007) regard it as the end of an adult's attempt to comprehend a child's understanding and the beginning of an adult interpretation of what has been revealed, which is often unmediated by children . By mitigation, Pollard (2004, 294)  Acknowledging the importance of visual methodologies when conducting research about children, the paper explores two specific tools -photo-elicitation and photo-voice. An interchangeable use of these terms appears in the literature , resulting in subsequent blurring of meaning; consequently, it necessitates a clearer rationale for their differences.

Photo-elicitation
The foundation of photo-elicitation is the idea of inserting one or more photographs into a research interview (Harper 2002;Bigante 2010) to generate verbal discussion (Thomas 2009). These visual images can be produced by either the researcher or the participant (Hatten, Forin and Adams 2013;Clark-Ibanez 2004). The study herein, defines photo-elicitation as a photograph produced by the researcher rather than participant; when the participant produces the photograph, the paper adopts Wang and Burris's (1997) terminology of photo-voice. Harper (2002) regards photo-elicitation as a tool that produces different kinds of information than more traditional forms of interviews, evoking feelings, memories and information. This is due to the parts of the brain that process visual information being older in evolutionary terms, than those processing verbal information. Consequently, one could regard it as a postmodern dialogue based on the authority of the participant rather than the researcher.
Drawing on Colliers' (1957) research, photo-elicitation sharpens the participants' memory and reduces the areas of misunderstanding. He notes that '[t]he pictures elicited longer and more comprehensive interviews but at the same time, helped subjects overcome the fatigue and repetition of conventional interviews' (Collier 1957, 858). This is of particular importance for younger children, since they have a right to expect that researchers will support them to communicate their perspectives by providing a rich array of resources and environments. It directly opposes perceptions of younger children as poor communicators and acknowledges that they can bring new insights and viewpoints (Bigante 2010) and create 'deep and interesting talk' (Harper 2002, 23 who make decisions over their lives' (Wang and Burris 1994, 172).
As with photo-elicitation, it acts as an intermediary in research, by enabling the researcher to focus the attention on the image rather than the participant. Grounded in Freire's (1970) empowerment education and education for critical consciousness through a focus on individual development, it brings about change through shared dialogue, and challenges documentary photography by regarding the participants as authorities in their own lives. Where it differs, is that it enables photographers to define the situations they see and represent them to others (Wang and Burris 1994;, through the process of taking the photographs. Its goals are threefold: to enable people to record and reflect upon their community's strengths and concerns; to promote critical dialogue and collective knowledge production; and to inform policy (Wang and Burris 1997). Kolb (2008) writes of four phases that are important within the use of photo-voice.
Firstly, the participant is asked a question and then to consider how to take photographs that reflect their viewpoint; this starts a cognitive process as they reflect on the meaning of the question. In the second or active phase, the participants implement their reflections by taking photographs of specific subjects in their social or material surroundings. The third or decoding phase, requires participants to consider their photographs and verbalise their thinking in an interview with the researcher, and if appropriate, in collaboration with other participants. The final phase involves researchers analysing the data -photographs, interviews, transcripts and observations -generated by the first three phases.
Research with younger age groups often adopts other visual research methods, such as drawings; however, use of photo-voice is less commonplace with this age group (Alaca, Rocca and Maggi 2017). It has the potential for empowering children, as the visual character of the method (in combination with interviews) makes abstract questions about difficult concepts, more approachable and accessible (Van Auken, Frisvoll and Stewart 2010).
Photographs can also help to sharpen children's abilities to reflect on, and explain, their experiences and perspectives; provide memory 'anchors' (Loeffler 2005); and gather valuable input from those who inhabit the environment (Prosser 2007).
Both visual methodological tools enable participants to reflect on the research question and discuss it with the researcher. The key difference is that with photo-voice the tacit knowledge about issues emerges as the participants go through a process of visualising the issue and producing images. The process encourages participant ownership of the question and engagement with, and empowerment in, the research process, through the narration that occurs, and enables the participants to become researchers in their own cultures and lives.

Methodology and research design
The study herein adopted a qualitative methodological approach operating within a constructivist and interpretivist paradigm, where categories and meaning are socially constructed (Bryman 2012). Ethical issues of consent, assent and dissent (Dockett, Perry and Kearney 2012) were duly considered.
Three schools engaged in the research: Riverside Infants (RI), Oak Ridge Primary (ORP) and Greenside Primary (GP). All children in the Reception classes (aged four to five years) were invited to undertake the research, however, only data from children whose parents consented were analysed -19 children from RI and 21 from ORP using photo-elicitation (PE), and 16 children from GP employing photo-voice (PV). To ensure anonymity, the application of pseudonyms for children and schools occurred. Data collection ensued over a six-week period, which facilitated the researcher's relationship with the children, enabling them to feel more at ease when engaging in conversations. The paper draws on data from twenty-four children using photo-elicitation and five children using photo-voice.

Research tools
This paper focuses on one element of a larger research project that employed a range of tools to understand children's perceptions of inclusion (Shaw, Messiou and Voutsina 2019). Acknowledging that inclusion is a difficult concept, it was explained in terms of places or activities where the children felt happy (included) or unhappy (not included).

Group interviews
Group interviews took place in each of the schools with four children at a time; these occurred after four weeks of observation. In RI and ORP, photo-elicitation was used, which required the researcher to take photographs and use them as an initial prompt to determine    6). Its strength lies in the premise that when people rank items, they are required to make explicit, the rationale for how they are organised through the process of discussion, reflection and negotiation with other group members (Clark 2012). Figure 6. Organisation of diamond ranking Due to the young age of the children, photo-elicitation was employed in a more structured way within the diamond ranking activity, than with other researchers (Harper 2002); nonetheless, it still provided a valuable means by which to stimulate thinking and support the children's agency (Niemi, Kumpulainen and Lipponen 2015). Following a pilot study with four and five year olds, in which the traditional nine items were used, the children were unable to rank them since this required a more nuanced method of categorisation and segregation, and they struggled to focus on the task for a longer time period. Consequently, only five images were utilised in this project, which were selected by the researcher.
In GP, the researcher adopted photo-voice and asked c hildren to take photographs of areas in the classroom and outdoor learning environments , where they felt, or did not feel included . D igital cameras provided flexibility in how the children chose to use them, whilst placing less emphasis on verbal communication . A digital recorder documented all interviews.

Data analysis
All interviews were transcribed and observational data and facial expressions and body language noted. The researcher adopted an inductive approach to coding, which allowed the theory to emerge from the data (Strauss and Corbin 1998), by considering the frequent, dominant, or significant themes inherent in raw data, without the restraints imposed by structured methodologies (Thomas 2006).
Analysis of data from the photo-elicitation and photo-voice tools, enabled consideration of the ways in which each tool allowed space for children to express themselves. Four strengths of visual tools emerged: Less reliance on or al competency ; Reducing power differentials; Enabling different voices to be heard; and Potential disclosure of otherwise hidden perceptions. Three limitations of visual tools also emerged: Not understanding the task/easily distracted/unable to answer; Influenced by peers; and Forgetting the task.

Findings
Verbatim quotes from children at RI and ORP using photo-elicitation (PE), and quotes and photographs taken by children at GP using photo-voice (PV), are used to illustrate the strengths and limitations of adopting visual methodologies with younger children.

Less reliance on oral competency
Qualitative research has a tendency to focus on methods that require a degree of oral literacy, which can present difficulties when working with younger children. One advantage of using Probing further, the researcher asked her whether she felt more included when she was learning with her friends or on her own. She replied with similar wording -' I like learning with Scarlet and Olivia.'

Reducing power differentials
Employing visual methodologies has the ability to empower children within the data collection process. Taking control of the camera in the photo-voice activities, enabled the process to become a responsive, child-led activity rather than a controlled, adult-led one.
Tommy (GP PV), a child with autism, appeared not to be interested at first, however later in the day, he grabbed the camera, ran immediately into the classroom and photographed a train set. The immediacy of using a digital camera enabled Tommy to present the image and discuss his reasoning for capturing the image directly afterwards. Photo-elicitation also enabled the children to take charge of the task. Emma (RI PE), for instance, elected not to answer questions about her experience of inclusion in school as asked, but rather she began discussing how she loves to colour pictures of her mother and then continued providing further details: ' Mummy's got short, short hair' ; ' Yeah, cos I never want to get it short like mummy's cos I'll look like a boy .' Whilst Emma may not have proffered information relating to the research itself, the flexibility of the task ensured that she was not constrained and inhibited by the research tool.

Concurring voices
Both photo-elicitation and photo-voice enabled children to express assenting points of view.
Some children perceived themselves as included when they were working with the teacher. concurred 'Yeah that's boring' and ventured further stating that ' All of them are boring. '

Divergent voices
Some children, however, presented opposing views of the same activity. Layla (ORP PE) perceived herself as included when working with the teacher because she 'loves writing …

Limitations of visual tools
Whilst this paper presents evidence to support the notion that visual methodological tools are powerful in opening up space and time for younger children to express themselves, they are not without difficulty.

Not understanding the task/easily distracted/unable to answer
Given the young age of the children herein, it may not be surprising to note that some did not

Influenced by peers
Whilst the intention of using the tools was to develop more participatory approaches to research, this appears compromised by the potential influence some children had over their

Forgetting the task
A further challenge specifically related to employing photo-voice with younger children, is that they may forget what has been asked of them. In this study, cameras were initially left with the class teacher and the children were informed that they could use them at any point during the week; unfortunately, the teachers did not remind the children and the children forgot they were available. However, when the researcher returned the following week and prompted the children to use the cameras, they provided a sufficiently effective means of capturing their interest and attention that enabled deep and meaningful discussions.

Discussion
Visual research methodologies have played a minor theoretical role in social research, because it has been a 'word-based' discipline, and the capacity of images to reveal 'the truth' has been questioned (Harper 2002, 17). Much research about children is with those of primary and secondary age, and is undertaken in settings where they are ready-made samples; these 'captive audiences' are often in schools. Here, children are least able to exercise participation rights since the balance of power is heavily skewed towards adults. Moreover, power differentials can be emphasised if adult accounts are filtered through adult constructions of the world, which may bear little or no relevance to children (Moore and Sixsmith 2000). Visual methodologies, however, can break down disparities in power and status between the researcher and the participant and enhance the richness of data. In adopting photo-voice interviews, for example, some of the power dynamics involved with regular interviews can be disrupted, by allowing the child to actively and subjectively interpret their photographic images, and in photo-elicitation the children can re-position the questioning and narrative in whatever way they choose.
The findings also question judgements about children's competence, which is often used as a principal obstacle blocking the empowerment of children as active researchers. Notwithstanding the recognition of younger children as active and competent agents in the process, such as facilitating them to collect data through photo-voice, researchers are still required to adopt a carefully crafted approach with a critical ethical or political praxis, rather than following a pre-prescribed set of techniques. Simply creating images and asking children to comment, or enabling children to create them, does not overcome the many possibilities that influence the production of images or, like all other texts, eradicate issues with interpretation. Similarly, all verbal interpretation of images will necessarily be positioned and reveal selective representations (Komulainen 2007 Emma) are empowered to reveal perspectives that do not necessarily present a positive narrative of their educational environment.

Conclusion
The study herein responds to the notion that younger children's voices are absent from empirical research (see Driessnack and Furukawa 2011;Gray and Winter 2011), by exploring participatory ways of conducting research, such as photo-elicitation and photo-voice, with a younger age group. It emphasises the potential that visual methodological tools, o perating within creative parameters that engage with less adult-centric and formalised approaches, afford to gaining new insights and viewpoints from younger children, through deep and interesting talk. Furthermore, it strengthens Rinaldi's (2006, 98) call for 'pedagogical research', and emphasises a culture of research with children that supports them to reflect and clarify their thinking about issues in their everyday lives.
Finally, the findings add value to existing research with younger children using photo-voice (Clark and Moss 2005;Cook and Hess 2007;Einarsdottir 2005;Linklater 2006), and bring new understanding of the benefits of employing photo-elicitation with younger children. The paper presents evidence that encourages educational research communities and practitioners to adopt visual methodologies that authentically capture and listen to the languages of younger children. Equally important, is the requirement for researchers and those that work with younger children, to think differently about how they can engage with children's voices through careful, critical and conscientious application of these methodological tools. The data herein, provides a meaningful and powerful context in which to raise the profile of the voices of younger children, through their active involvement in research, which can consequently advance understandings of their perceptions of matters affecting them in their lives.