Access to Justice software development, Participatory Action Research Methods and Researching the Lived Experiences of British Military Veterans

Participatory action research (PAR) methods aim to position the people who are most affected by the issue being studied as equal partners in the research process through a cyclical process of data gathering, data analysis, planning and implementing action and evaluation and reflection. In doing so, it ensures that the research better reflects participants’ ideas, priorities, and needs, thereby enhancing its validity and relevance and the support for the findings and proposed changes. Furthermore, it generates immediately applicable results. In this paper, we reflect on our experiences of developing the UK’s first access to justice platform for veterans and their families through an ongoing PAR project that brought together armed forces veterans, representatives from veterans' service providers, and the Veterans Legal Link team members comprising of legal academics, lawyers, sociologists, computer software designers and graphic designers to collect, interpret, and apply community information to address issues related to the delivery of access to justice. We present findings from Stages 1 and 2 of our three-stage iterative research process which includes the following steps: Understanding and cross-checking the lived experience of the veteran community (Stage 1), developing and testing a prototype of the access to justice platform (Stage 2) and creating the final product and giving real users an opportunity to use the platform (Stage 3). Data collection and analysis from Stage 1 of the study informed the themes that underpinned Stage 2. Specifically, data was collected through the following methods: co-facilitated focus group discussions, a web survey that was codesigned with veteran community stakeholders and remote and digitally enabled ethnographic research methods. We include several reflections that may help legal practitioners and researchers interested in applying PAR within the area of access to justice and the field of legal research.


Introduction
As a form of applied research, participatory action research (PAR) 1 has been used extensively in many disciplines however it is comparatively rare in the field of law 2 and even less common in the area of access to justice research. 3 In this paper, we reflect on our experiences of developing the UK's first access to justice platform for veterans (former members of the British Armed forces who served for at least one day 4 ) and their families through an ongoing PAR project that brought together armed forces veterans, representatives from veterans' service providers, and the Veterans Legal Link (VLL) team members comprising of legal academics, lawyers, sociologists, computer software designers and graphic designers to collect, interpret, and apply community information to address issues related to the delivery of access to justice.
In doing so, our aims were to contribute to the small but growing literature on PAR in the field of law and to demonstrate the usefulness of PAR methodology to access to justice research projects. This Article proceeds as follows. Section 2 introduces the wider VLL project which had a catalytic and spin-off effect on the access to justice platform project thereby serving as a prelude to the subsequent sections. In Section 3, we unpack the methodological and epistemological foundations of PAR, discuss the origins of PAR and compare PAR and conventional research 5 thereby providing the groundwork for Section 4. In Section 4, we present findings from Stages 1 and 2 of our three-stage iterative research process which incorporates the following steps: Understanding and cross-checking the lived experience of the veteran community (Stage 1), developing and testing a prototype of the access to justice platform (Stage 2) and creating the final product and giving users an opportunity to use the platform (Stage 3). As elaborated in more detail below, data collection and analysis from Stage 1 of the study informed the themes that underpinned Stage 2. Specifically, data was collected through the following methods: co-facilitated focus group discussions, a web survey that was codesigned with veteran community stakeholders and remote and digitally enabled ethnographic research methods. Finally, in Section 5, we discuss the benefits and challenges of PAR, while also offering several reflections that may help legal practitioners and researchers interested in applying PAR within the area of access to justice and the field of legal research.

The Veterans Legal Link project
The Veterans Legal Link (VLL) project is an access to justice project that provides free legal advice services and professional signposting for veterans and their families. The VLL's services were accessible through the use of drop-in centres across Wales, as well as being accessible via phone and email. The VLL's services are available to any British military or blue-light veterans and their families. The VLL grew out of the Principal Investigator's research interests and was established in 2015 in response to the introduction of the Legal Aid, Sentencing and Punishment of Offenders Act 2012 (LASPO). LASPO was introduced as part of the government's programme of spending cuts to achieve significant savings to the legal aid budget. However, LASPO has had a disproportionate impact on veterans and other vulnerable populations. For instance, the Amnesty International report "Cuts that Hurt" 6 observed that LASPO has had an unequal impact on people with additional vulnerabilities and or disadvantages that make accessing, navigating and understanding the legal process harder. This includes those with mental illnesses, low numeracy and literacy levels, and alcohol and drug conditions. 7 The VLL project had been in existence for six years-the minimum recommended number for a long-standing academic-community partnership 8 . In 2019, we conducted a usability evaluation (as there was a continuing trend of our service users accessing our service through routes other than the drop-in centres) to better understand VLL's service users and their needs and discovered that only 30% of the service users were accessing the service via the drop-in clinics. This presented the VLL with an opportunity to begin research into the optimal mode of delivery for the service on an ongoing basis. This started our journey using PAR as our methodology in order to, "make sure the questions asked and methods used do justice to the pressing issues at hand, the richness of participant knowledge and local views about the matters under investigation." 9 The first phase study involved understanding and cross-checking the lived experience of the veteran community. We carried out an initial consultation in July 2019 with multiple veteran organisations and the wider veterans' community to better understand their lived experience when accessing services and the wider needs of the community (addressed in greater detail in section 4 on the research process). The initial consultation indicated a need for an additional complementary online provision for the delivery of the 6 Amnesty International, Cuts that hurt: the impact of legal aid cuts in England on access to justice (2016 VLL's services. This crystallized into the access to justice platform project. Phase 2 involved the creation of an online mobile and web-based portal for the delivery of the service and was greeted with enthusiasm and positivity by the veterans and organisations consulted in the initial stage. The access to justice platform project was led by the VLL and guided by a steering group composed of armed forces veterans, representatives from veterans' service providers (both public and private) and veteran organisations. The Veterans Legal Link (VLL) team was comprised of legal academics, lawyers, sociologists, computer software designers and graphic designers and aimed to collect, interpret, and apply community information to address issues related to the delivery of access to justice. From its very beginnings, the research methodology into the veterans, community was firmly rooted in the PAR approach.

Participatory Action Research (PAR)
PAR encapsulates an epistemological position 10 , a research methodology, 11 and a process for PAR focuses on social change that promotes democracy and challenges inequality; is context-specific, often targeted on the needs of a particular group; is an iterative cycle of research, action and reflection; and often seeks to 'liberate' participants to have a greater awareness of their situation in order to take action". 16 One of the aims of our study was to challenge the inequality in access to justice for veterans and their families living on a low income in rural and remote communities of the UK.
We employed an iterative methodology 17 underpinned by a continuous cycle of data gathering, data analysis, formulation of an action plan, implementation of the action plan and evaluation ( Figure 1). Moreover, through the project steering group, composed of academic researchers and community partners, our PAR approach incorporated continuous dialogue, relationship building and active and genuine participation of veteran community stakeholders in the research process and thus had a liberating effect on them. 18 In addition, as "there is no one way to implement PAR" 19 we elected to use co-facilitated focus group discussions, a web survey that was codesigned with veteran community stakeholders and remote and digitally enabled ethnographic research methods to undertake our research at the appropriate phases (outlined in Figure 2). With the onset of Covid-19 it was particularly important to use a digitally enabled method for phase 2.

Figure 2: A three-stage iterative process for developing the access to justice platform
Drawing on the experiences of researchers from other disciplines, it is evident that PAR as a method of research has both benefits and challenges, like all research methods. The strength of PAR as a research methodology stems from various aspects that, within the context of the research we are conducting, produces knowledge that can be applied directly to the local context and to democratise the coproduction of knowledge by collaborating with those most directly affected by the research. 20 The veterans' community, particularly in Wales (the primary geographic region of VLL service users), is characterised by few major metropolitan centres however most of the population live in rural, geographically isolated and lower income areas. 21 As a result, the local knowledge is essential to accurate understanding of problems and the development of effective interventions best suited to provide access to justice for an underrepresented community group and by engaging them through participation in the research process the resulting knowledge is embedded in local contexts. 22 Furthermore, by engaging with the veterans' community in a participatory way we were able to "ensure the relevancy of research questions; increase the capacity of data collection, analysis, and interpretation… and enhance program recruitment, sustainability, and extension". 23 This approach also provided a means for engendering trust 24 and building of community relationships with a community that has a deep mistrust of civilians 25 (discussed in greater detail in section 5. and cultural sensitivity in data gathering). 26 On the other hand, the challenges posed by PAR included participant recruitment particularly when carried out online 27 and the desire for a high level of community involvement. 28 Through our study, we saw first-hand some of the abovementioned benefits and challenges and discuss them in more detail later.

The Research Process
In this paper we reflect on an access to justice platform project which originated from the VLL, a well-established partnership between academics and community partners. At the outset, we should point out that our research project is still ongoing and hence, we present findings from Stages 1 and 2 of our three-stage iterative research process. Data collection and analysis from Stage 1 of the study informed the themes that underpinned Stage 2. Throughout the lifetime of the project, and since its inception, ethics approval for the overall VLL project and for the specific research activities have been granted through Aberystwyth Universities Ethics Board.
In phase 1, data was collected through a combination of co-facilitated focus groups and a web survey. In collaboration with existing community partners we conducted Focus Group Discussions (FGD), we used convenience sampling to generate a heterogeneous sample of veterans (older and younger, men and women, service users and service providers, individuals with varying degrees of technical literacy and geographic regions (North, South, East and West)) and sent participants a copy of the information sheet and a consent form by email. with between 5-10 participants and aimed to collect data surrounding the following themes, veterans use of technology, barriers to accessing online services, perspectives and utility of current services and additional unmet needs for the veteran community. The focus group participants were recruited from the steering groups' service users and were co-facilitated by their organisations (RBL and Change Step) in conjunction with VLL. The co-facilitation was needed due to the high levels of mistrust among veterans towards civilians and were held both digitally (4) and in-person (1). Following Breen's recommendations 31 , we devoted the lion's share of our discussion time to probing participants' experiences of using technology and accessing online services, asking them to share and compare their experiences, and discussing the extent to which they agree or disagree with each other. It was not until the final third of the focus groups that we explicitly asked the following questions and facilitated discussion on the topics: • How do participants use technology currently?; • What are the barriers to accessing online services?; • What is the response to a proposed enhanced online access to justice portal and online  Furthermore, we used NVivo 12 software for the inductive coding, analysis and interpretation of the open-ended survey question data and followed the following six-step thematic framework recommended by Braun and Clarke 36 to review critically the data and to environment for conducting prototype testing with users. Participants of this stage of the research were provided detailed information sheets and informed consent with a variety of options to address security and privacy concerns such as the ability to op-in/out of screen sharing, allowing them to opt for a camera off session and to not have the sessions recorded.
All of the participants in this stage agreed to screen sharing, camera on, and to have the sessions recorded.

Reflections on the application of PAR for the development of an access to justice platform
In the following subsections, we reflect on some of the benefits that accrued from our efforts to apply PAR for the development of an access to justice platform. These benefits are as follows: enhance recruitment capacity, ensure ecological and cultural sensitivity in data gathering and instrumental benefits. On the other hand, we also reflect on the following challenges that we faced as legal academics engaging in a sustained PAR project: deception by online survey participants and involvement of community stakeholders in some but not all phases of the research process.

Enhance recruitment capacity
One of the benefits of PAR is partnership synergy, that is the "combined effect of complementary tangible and intangible partnership assets and enabling processes that gives partnerships unique advantages over the work of individual people or organizations working towards the same goals". 43 Notably, in our project, recruitment of community members to the steering group was accelerated by the fact that academic-community partners had a longstanding relationship dating back more than 6 years via the VLL project. Thus, existing partners who were representatives of respected veteran community organisations readily accepted the invitation to join the project steering group and hence actively participated in meetings and were strong believers in the project's benefits for the communities they served. Furthermore, from a synergistic perspective, the fact that the project had forged strong links with respected and trusted veterans' community organisations proved critical in the recruitment of additional community representatives to the steering group and the recruitment of veteran participants into our research. Thus, the involvement of community partners enhanced the credibility of the project, increased trust and reduced barriers to recruitment.

Ensure ecological and cultural sensitivity in data gathering
As civilian academic researchers, we were conscious of the fact that for many veterans transitioning from service in the armed forces to civilian life, developing trusting relationships with "civilian outsiders" could be a difficult proposition, as the military is a close-knit community. 44 In our project, through their familiarity with and sensitivity to access to justice issues, veteran community partners who were members of the project steering group were able to ensure ecological and cultural sensitivity in data gathering. Specifically, they contributed academic researchers to veteran community members. They were also involved in the development of survey questions and co-facilitated all the focus groups. For example, the fact that the total number of valid surveys completed through the duration of our survey was 528 serves as evidence for the cultural acceptability of the survey 45 . Thus, the contribution of veteran community partners was critical to the acceptability and validity of our research.

Instrumental benefits
According to Hagan "[m]ost access-to-justice technologies are designed by lawyers and reflect lawyers' perspectives on what people need. Hence it is not surprising that most of these technologies do not fulfil their promise because the people they are designed to serve do not use them". 46 Our own experience of applying PAR to the generation of ideas for the development of an access to justice platform validates Hagan's statement. It was apparent from our research that veterans were acutely aware of problems with the existing approach to delivering access to justice and as "experts by experience" including them in key phases of the research process helped to facilitate new concept generation for the design and development of the access to justice platform. This collaborative approach has direct implications for the acceptability, uptake and adoption of the platform by veteran end users as the platform is more likely to meet their expectations and requirements. However, as mentioned above, we have completed Stages 1 and 2 of our three-stage iterative research process which comprises the following steps: Understanding and cross-checking the lived experience of the veteran community (Stage 1), developing and testing a prototype of the access to justice platform (Stage 2) and creating the final product and giving real users an opportunity to use the platform 45 We conducted a qualitative analysis of the survey results however were we to have done a quantitative statistical analysis of the data the sample size would have represented a 99% confidence level with between a 5-6% margin of error. 46 M Hagan, M 'Participatory design for innovation in access to justice' (2019) 148(1) Daedalus 120.
(Stage 3); and hence it is too early to draw any conclusions about the impact of PAR on the uptake and adoption of the access to justice platform.

Deception by online survey participants
The misrepresentation of survey participants who were eligible to take part in our research was one of the major challenges that we faced within the context of this project. We should first Notably, we offered an option to be included in a drawing of five £20 Amazon gift card as an inducement to participate. This incentive was advertised within the social media postings on both our LinkedIn and Facebook platforms. This, we suspect, was the main motivation for non-eligible participants to complete the survey to obtain the financial incentive. This became a larger problem when it became apparent that not only were we receiving individual noneligible responses but that some respondents were "spamming" (repeatedly filling out the survey) responses. The phenomena of receiving spam responses took several days from the launch of the survey to begin taking place and several additional days for it to be identified that these were potentially mis-represented responses to the questionnaire. Several aspects of the responses constituted a cause for concern. These included the following. First, some of the free text responses that were being received were in languages other than English or with grammatical structures that suggested a non-native English speaker. While this for some surveys may not be indicative of invalid responses, this questionnaire was intended for exforces personnel who served in the British armed forces whose language proficiency in English would either be to a native level or a Common European Framework of Reference (CERF) A1 level. Second, surveys were completed in succession over a period of time as indicated by the time stamp for the submission or were completed (from start to finish) in an unbelievably short amount of time, much faster than average reading speeds for the amount of text in the survey.
And third, there were duplicated or suspiciously similar responses across the questions as JISC Online Surveys do not prevent a survey from being completed many times on the same computer or from the same IP address unless survey access control is utilised. All of these posed threats to sample validity and data integrity. Therefore, in order to identify and minimize misrepresentation by participants seeking enrolment in our online survey thereby enabling sectional questionnaire'. (2020) 22 (10)  veterans to make their voices heard we used the following combination of strategies recommended in the literature: technical/software strategies and data analytic strategies. With regard to the former, we analysed the dataset to identify unusual or unexpected completion patterns. For example, we found that some respondents were completing the total survey of 18 items and 6 sub-items in 3 minutes or less, whereas the majority of respondents required 6 minutes. In terms of the latter, we determined if data from the sample of respondents who were suspected of misrepresenting their eligibility, differed significantly from the rest of the study sample or if results of the study were substantially different when either including or excluding their data from analyses. 50 Once identified suspected fraudulent results were removed from the sample, leaving 528 survey responses. These responses were the sample that informed the next phase of the PAR cycle.

Involvement of community stakeholders in some but not all phases of the research process
Ideally, PAR calls for the active involvement of community stakeholders as equal partners in all phases of the research process from defining relevant research questions, to planning, designing and implementing the investigation, strengthening recruitment strategies, collecting and analysing data and interpreting and applying findings and disseminating outcomes; based on our own experience these requirements are very difficult to meet. This is consistent with linked to what Hayward and colleagues refer to as "self-defined boundaries" 56 and that these boundaries could fluctuate depending on circumstances and contexts and hence it was necessary for us to be flexible and ready to adapt to possible changes during the research process.

Conclusion
In this paper, we reflected on our experiences of developing the UK's first access to justice platform for veterans and their families through an ongoing PAR project that brought together armed forces veterans, representatives from veterans' service providers, and the Veterans Legal Link team members comprising of legal academics, lawyers, sociologists, computer software designers and graphic designers to collect, interpret, and apply community information to address issues related to the delivery of access to justice. We presented findings from Stages 1 and 2 of our three-stage iterative research process which included the following steps: Understanding and cross-checking the lived experience of the veteran community (Stage 1), developing and testing a prototype of the access to justice platform (Stage 2) and creating the final product and giving real users an opportunity to use the platform (Stage 3). Data collection and analysis from Stage 1 of the study informed the themes that underpinned Stage 2. As demonstrated by the foregoing discussion, data was collected through the following methods: co-facilitated focus group discussions, a web survey that was codesigned with veteran community stakeholders and remote and digitally enabled ethnographic research methods. We included several reflections that may help legal practitioners and researchers interested in applying PAR within the area of access to justice and the field of legal research. As discussed above, some of the benefits that accrued from our efforts to apply PAR for the development of 56 See Hayward et al (n 54).