Introduction

The so-called ‘data revolution’ has had a major impact on the humanitarian sector over the past ten years and more (Meier, 2012; Pearn et al., 2022), not least due to its promise of quicker and more efficient evidence-based responses to situations of conflict and disaster. Yet as scholars have already emphasised, the development of data-driven humanitarianism also generates a series of problems and trade-offs. For example, the privacy of beneficiaries can conflict with an impulse to improve the efficiency of data-targeted aid, while sensitive data can be hard to identify and risk mitigation can be seen as secondary to decisive action (see Sandvik et al., 2014). Moreover, the proliferation of cross-sector partnerships of humanitarian organisations and tech companies can exacerbate asymmetrical humanitarian/beneficiary relations (Henriksen, 2023) while impeding efforts to enhance the ‘efficiency, effectiveness and greater accountability’ of humanitarian action (IASC, 2023). Indeed, new data modalities and practices of data sharing reconfigure the relationship between data collectors and data subjects, with the proliferation and complexity of contemporary humanitarian data rendering the management and use of data increasingly opaque. There thus remain serious problems surrounding the ethics and practice of data-driven humanitarianism.

How then might more ethical data practices be introduced into the field of humanitarian practice? This article reflects on the potential and limitations of critical data literacy as a transformative method in the context of the datafication of the humanitarian sector. It draws on research carried out with internally displaced persons (IDPs) in Nigeria and South Sudan during 2020–23 as part of a collaborative international project examining the practice and ethics of data collection and use. Our research suggests that the datafication of the humanitarian sector generates distance between humanitarians and affected communities (see Duffield, 2018) while exacerbating the drive toward extractive relations and processes that entrench power asymmetries and longer-standing relations of colonial dependency and dispossession (see Aradau and Tazzioli, 2020; Gray, 2023; Lemberg-Pedersen and Haioty, 2020; Madianou, 2019; Squire and Alozie, 2023). Datafication deepens communication barriers between those providing aid and affected communities of IDPs (Jones and Logo, in progress), further disempowering ‘beneficiaries’ through the very development of data-driven humanitarianism. The article discusses the participatory ethos that the Data and Displacement project developed with the aim of countering some of these tendencies. This involved ongoing engagement with IDPs over time through the co-production of data-literacy tools. Reflecting on the significance of our findings for humanitarian practitioners as well as for academics working in the field of humanitarianism and displacement, we make the case for a collective commitment to the meaningful engagement of affected communities through co-production and shared reflection on the production and use of data. We also caution against viewing data literacy as an easy fix to empowerment challenges, both in the conduct of humanitarian work and in the implementation of research.

Critical Data Literacy

The approach to data literacy that is developed in this article draws inspiration from critical scholarship in the field of data studies (Dalton et al., 2016). It is grounded in an appreciation of the ‘social life of data’, whereby data are understood as inseparable from and, indeed, constitutive of the ongoing formation of social relations and power dynamics (Ruppert and Scheel, 2021). From this perspective, data are not regarded simply as consisting of digital representations of existing phenomena: rather, there is appreciation of the need to consider digital artefacts as they are embedded in the materiality of ‘data practices’ (Bates et al., 2016; Dalton and Thatcher, 2014; Tkacz et al., 2021). Understanding the role of data in this sense requires that we pay close attention to the socio-material practices in which specific digital technologies and datasets are generated, processed, circulated and used – alongside their governance and emerging power relations (Porto de Albuquerque et al., 2021).

This article also draws on scholarship in the field of humanitarianism, which pays attention to the hierarchical relations between beneficiaries of aid and those providing assistance (Fassin, 2012; Madianou, 2019). In terms of data, this entails thinking critically about the asymmetries that are embedded within the relations through which people engage with data and digital tools, particularly those that exist between humanitarian actors and marginalised groups of people such as those in situations of internal displacement. These asymmetries impose significant challenges regarding how to meaningfully engage affected communities – not for an extractive and instrumental ‘sensing’ of data that others will make sense of, but in terms of transformative modes of engagement with and through data processes (Porto de Albuquerque and Almeida, 2020).

By bridging scholarship across the fields of data studies and humanitarian studies, we seek to deepen appreciation of the importance of IDP data literacies to the development of effective and ethical practices of data-driven humanitarian assistance. This advances a stream of studies on critical data literacy (D’Ignazio and Bhargava, 2015; Fotopoulou, 2021; Sander, 2020; Tygel and Kirsch, 2016). Importantly, however, and in line with Gray et al. (2018), we seek to move beyond an instrumental conception of data literacy as consisting only of acquiring the technical, numerical and statistical capacities and skills to use digital tools to data. Rather than de-contextualised and universal technical skills, data literacies should be viewed in relation to other types of literacies (e.g. written, media, information, digital literacies) (Carmi et al., 2020), as well as in relation to the multiple real-life material conditions and social contexts within which data practices acquire their meaning (Fotopoulou, 2021). Our approach to critical data literacy thus recognises that the multiplicity and situatedness of data literacies requires being attentive to the changing societal needs associated with these, as well as being responsive to new data practices and technical procedures for data collection, processing and usage (Yates and Carmi, 2020). This implies a critical perspective of how data practices can generate biases that not only reflect but also reproduce structural social inequalities and unequal power relationships (Eubanks, 2018).

Developing data literacy from this critical perspective necessitates creative tactics (D’Ignazio, 2017), and ways of engaging communities through data in terms that are specific to their situations and responsive to their worldviews and particular circumstances. Drawing inspiration from the critical pedagogy of Freire (1970), our approach examines practices of reading and writing data alongside the capacity of ‘reading and writing the world’. In other words, we examine data practices as embedded within processes of knowledge co-production, with the aim of enhancing critical consciousness surrounding the existential, concrete and frequently oppressive situations of marginalised people (Tygel and Kirsch, 2016; Porto de Albuquerque et al., 2023). The need for data literacy to engender such awareness and critical reflection has been emphasised by a number of current works (Gray et al., 2018; Sander, 2020). However, a question remains as to how far critical data literacies can be developed to actively respond to ongoing trends of datafication and data-driven epistemologies in different sectors, including by reducing the intensification of social injustice (Pangrazio and Sefton-Green, 2020). Reflecting on such concerns in the humanitarian context, our starting point is not that IDPs merely need to learn more about how their data is generated and used, but rather that there is a need for broader critical reflection on the communication barriers and epistemic injustices that are exacerbated by data-driven humanitarianism (Squire and Alozie, 2023; Jones and Logo, in progress). We therefore consider the transformative potential and limits of advancing critical data literacies through processes of data production and use within the humanitarian sector.

Data and Displacement

This article draws on research carried out as part of a collaborative international project examining the practice and ethics of IDP data collection and use. The Data and Displacement project ran from 2020–23, and involved a collaboration of scholars from Nigeria, South Sudan and the United Kingdom. Our research brings together insights from multiple disciplines, including Data Science, Geography, International Relations, Law, Politics, Peace Studies and Sociology, and we worked in partnership with the International Organization for Migration (IOM) throughout the project. As well as undertaking interviews with humanitarians specialised in data and information management, we also undertook research with IDPs and stakeholders in camp-like settings across north-eastern Nigeria and South Sudan. These locations were chosen as conflict regions whereby the collection of large-scale data is well-established and basic provisions are distributed using e-ID cards, with regular needs assessments and baseline data produced by international agencies and NGOs in both contexts.

Internal displacement has been a significant issue in the north-eastern region of Nigeria for the past two decades, where the activities of non-state armed groups have generated a range of challenges for the 2.2 million internally displaced in the country. Based on data from the UN Office for the Coordination of Humanitarian Affairs (OCHA), no less than 8.7 million of the 13.1 million population in insurgency areas of north-eastern Nigeria have been identified as requiring humanitarian assistance as of 2021, including 5 million children, 1.74 million women and 1.4 million disabled people (OCHA, 2022a). Although South Sudan gained independence on 9 July 2011, conflict broke out in December 2013 leading to high levels of violence and displacement with over 2 million internally displaced persons within South Sudan (OCHA, 2022b). As of July 2022, there were five camps in South Sudan, which in 2020 were re-designated from Protection of Civilian Sites (PoCs) managed by the United Nations Mission in South Sudan to IDP camps under the management of the South Sudanese Government.

From 2021 to 2022, the Data and Displacement research team conducted in-depth qualitative interviews in north-eastern Nigeria with fifty IDPs across five camps in Maiduguri, Borno. We also interviewed twenty stakeholders and practitioners working with IDPs in the region to deepen our understanding of data-driven humanitarianism in contexts of internal displacement. A total of twenty-two interviews were carried out in phase one, from March to April 2021, eleven with IDPs and eleven with stakeholders. A further forty-eight interviews were carried out from October 2021 to January 2022, thirty-nine with IDPs and nine with stakeholders. Since our research was conducted, there have been increased moves to resettle IDPs and close camp-like settings within Nigeria. However, as of July 2022, the IDP population within camps and camp-like settings remained at 988,428, or 40 per cent of the IDP population in north-eastern Nigeria (IOM, 2022a). Our research, therefore, provides important insights into the impact of data-driven humanitarianism for IDPs in the region and is also of broader interest across a range of sites of ongoing conflict and displacement.

In South Sudan, the team conducted interviews in Juba and Bentiu IDP camps. As of April 2022, over 135,000 IDPs were registered in the Bentiu camp (IOM, 2022b), while in January 2022 approximately 33,000 IDPs were registered in the Juba camp. There were twelve humanitarian service clusters in the camps, comprising a range of international and local humanitarian organisations that provide services such as camp management, water, sanitation and hygiene, health, psychosocial support, food, shelter, education and protection. From 2021 to 2022 we conducted in-depth interviews with fifty IDPs across the two camps and interviewed twenty stakeholders and practitioners working with IDPs in the camps. All our interviews were semi-structured and were undertaken face-to-face in two phases to enable a review of initial interview data and questions. On average, interviews lasted around one hour, with questions for IDPs focused on displacement experiences, camp conditions, data collection and use, and visual materials (maps and project infographics). Flexibility was maintained in the questioning to allow research participants to elaborate on areas of concern or to raise new issues, as relevant to their experience and expertise. IDPs were recruited to ensure a diverse sample of gender and age. Members of our research team recruited research participants directly.

In undertaking in-depth interviews with IDPs and stakeholders in camps, we engaged a series of visual tools to facilitate engagement with our research participants. As well as using local maps of the camp, we developed a simplified infographic to render as accessible as possible complex data processes or the ‘data journey’ (Bates et al., 2016). Using the infographic in our interviews with both IDPs and stakeholders enhanced our understanding of IDP perspectives as a research team (see Squire et al., 2022: 28), allowing us both to centre the lived experiences of IDPs in our discussions, as well as to assess understandings of data, data collection and use on the part of our research participants. Discussion of the visualisation with IDPs, stakeholders and within our trandisciplinary team also enabled us to move beyond an initial idealised representation of current data practices in linear form. On this basis, we developed a complex representation of the ‘data journey’ to highlight the complex entwinement of different processes and the frictions that these generate (Bates, 2018). This was used to engage stakeholders in dissemination workshops in Abuja, Juba and Bentiu, where we gathered further reflections and inputs from IDPs and humanitarian practitioners. We also engaged in reflective sessions with selected IDPs and practitioners in Abuja, which enabled further refinement of the infographic (see Squire et al., 2022: 29).

Throughout the duration of the project, we sought to develop a participatory ethos across the research team (Squire et al., 2021), with the aim of transforming the ‘unequal power dynamics that all too often underpin the design, collection, use and governance of data’ in the humanitarian sector (Data Values Project, 2022: 6), as well as in academic research (Chonka et al., 2022). This culminated with a co-production workshop organised in Lagos in January 2023, where we designed and recorded a series of digital clips focused on data rights with IDPs directly. We will discuss the challenges as well as the benefits of such processes later in this article. For now, we want to turn to the project findings to highlight the communicative barriers and challenges that we found to be exacerbated through the datafication of the humanitarian sector.

Communication Barriers

Our analysis suggests that the datafication of the humanitarian sector exacerbates the drive toward extractive relations and processes while also deepening communication barriers between affected communities of IDPs and those responsible for the provision of aid (Squire and Alozie, 2023; Jones and Logo, in progress). It also highlights a series of ethical concerns surrounding the accuracy of IDP data in each context (Squire et al., 2022). For example, in north-eastern Nigeria we found discrepancies in definitions of vulnerability used in contexts of data gathering; coordination and reliability issues across institutions and actors; and inconsistencies in the implementation of ethics processes of informed consent and transparency surrounding data utilisation (Fayehun and Àkànle, 2022). Many IDPs in this context did not understand why their data was being gathered or how it was being managed, and overall data literacy levels appeared to be low:

We don’t know the place where they use to share our information, just they say they take our data to take to their organisation. (Female IDP, Nigeria, NIG.IDP.14)

No, they are not telling us, just that when they collected information from us, they are going and that we would not see them again. (Male IDP, Nigeria, NIG.IDP.04)

Insufficient data ethics training was also evident among humanitarian data actors, as well as significant gaps in the personnel, technological and infrastructural resources required in data storage and handling processes. This is reflected in the limitations of informed consent reported by many IDPs:

No, they are not giving us any consent form to sign … They will just start the interview. (Female IDP, Nigeria, NIG.IDP.03)

Difficulties in ensuring accurate data gathering, storing, retrieval and utilisation in this regard not only pose problems for IDPs but also for the development of sustainable humanitarian practice and interventions (Fayehun et al., 2023).

It is worth noting that general literacy levels are relatively good in Nigeria in comparison to those in South Sudan yet despite this, consent processes are largely not understood by displaced persons. This is all the more problematic given that IDPs are subject to repeated data collection processes – one of our research participants in Abuja told us in one of the project workshops that he provides data as much as ‘five times a day’ to different data collectors. In South Sudan, IDPs had both low levels of data literacy as well as low levels of literacy in general:

We don’t ask because we don’t know the system. They didn’t explain it to us. (Female IDP, SS.IDP.08)

Those that are very vulnerable, some are, they don’t speak up, they don’t have voice. (NGO Representative, SS.SH.09)

In South Sudan, we found that the recent role of international agencies in the management of camps and provision of aid led to greater consistency in data collection and use than in the Nigerian context. For example, interviews with stakeholders suggest that international humanitarian actors, especially UN agencies, adhere to elaborate data protocols and processes compared to their national or local counterparts. That can be due to the funds and resources international organisations mobilise for data processing and protocols, which local or national organisations often do not have (Male IDP, SS.02.SH.2). Nevertheless, also notable was the lack of understanding of data processes on the part of IDPs and a sense that data was very much extracted from IDPs without meaningful follow-up:

[O]rganisations can come and take information and do not return and sometimes people can come and take the information and we do not know what they will do with the information and sometimes we are afraid. (Male IDP, SS.01.IDP.01)

Some humanitarian practitioners in South Sudan told us that they ensure language assistants lead the process when data is being taken, to enhance the understanding of IDPs. They also suggested that ample time is taken to explain difficult questions and what is expected of IDPs and to clarify why a particular dataset is taken and how it will be used. We were also told that feedback is provided in various ways through IDP structures to cascade information to everyone in the IDP camps. However, both humanitarians and IDPs in Bentiu and Juba provided a more mixed picture, either disagreeing that data protocols were being followed or showing a lack of awareness of these:

There is very little information shared back with IDPs – I have heard them complain that they say they are tired with the data taking that occurs all the time in the camp! The challenge is that all of the agencies and NGOs are here to execute their mandates. There are very many IDPs and while some agencies attempt to adhere to data protection and ethics of data collection, I think that if there is no supervision of data taking, the data taken will have gaps. (NGO Representative, SS. 02.SH.12)

[T]he humanitarians take the information to [the] funder but … they don’t give feedback to us and explain to us that this is what happened to us to the data we have [given]. (Male IDP, SS.02 IDP.28)

Because we trust them, we have never questioned their intentions and even when they collect data from us, we do not ask questions. Today we learnt that we have rights and that it is okay for us to ask [humanitarians] questions. (IDP speaking at Bentiu workshop 17.08.22)

Challenges to Engagement

Intervening in contexts whereby data collection and use is already a tense issue raises particular challenges and responsibilities. Our aim as researchers was to engage our interviewees within the research process as participants rather than simply as respondents. Our methods were designed with this in mind: we combined semi-structured interviews with limited participant observation while creating infographics that we hoped IDPs would both respond to and help shape (Squire et al., 2021). Motivating this approach, as indicated above, was our concern to generate a space within which the affected communities could speak for themselves and be empowered to reflect on issues of data rights and data justice. However, we also faced significant dilemmas in operationalising these methods, which arguably reflect some of the dilemmas which humanitarian practitioners face when collecting, processing and using data. In conflict and post-conflict contexts, where conflict can occur unexpectedly, humanitarian actors often fail to meet the basic needs of a population fleeing harm. In such contexts, data processes often remain substandard as basic needs are prioritised over IDP engagement (see also Sandvik et al., 2014). In our project, we had financial and time pressures and we were conducting interviews in contexts of extreme deprivation. This meant that the gap between our concerns and those of our research participants was stark, while the time and resources we needed to address this were limited.

What became clear in the first phase of fieldwork in both contexts was that even our simple infographic was not easy to understand for many of the IDPs who we interviewed. This was most marked in South Sudan, where our research participants were reluctant to discuss the visual materials, did not seem to understand their purpose, and could not easily translate our research into something which was meaningful to them. Some in Juba were even disinterested in the visual materials. This was partly due to their time constraints, finding it difficult to spend much time with us while aware of caring responsibilities and pressures to find paid work. There was also an apparent inability to interpret the infographic. In the second phase of fieldwork, we were more experienced as a team in discussing the infographics and found a wider range of responses to the visual materials. In South Sudan, a few IDPs in Bentiu were able to engage effectively with a map of the camp and with the simple ‘data journey’ that we had developed (e.g. by identifying key data processes and areas of safety and insecurity in the camp). There were also moments when eyes lit up when the language assistant pointed to their residential blocks and the facilities housing humanitarian actors. In Nigeria, many of our research participants expressed appreciation for the information provided by the map and infographic.

The methods that we had hoped would be the most effective in a participatory sense were not always effective, however, particularly in South Sudan where a more generalised sense of data disempowerment prevailed and literacies were more generally low. IDPs in South Sudan were primarily concerned with their safety and with having their basic needs met and so faced additional challenges in demanding greater empowerment within the data production process. For example, some did not want to risk alienating the very people who were able to help them (i.e. humanitarians). Many who we interviewed in South Sudan felt that they owed their very existence and subsistence to humanitarian actors and that asking questions may be misinterpreted as being unappreciative of humanitarian assistance received over the years.¹ Data disempowerment or disengagement in this regard might be understood as generated by reliance on humanitarians, difficult living conditions, relatively low literacy levels, as well as limited resources among humanitarian organisations.

In South Sudan in particular, there was often limited demand for feedback and relatively high levels of acceptance of data collection processes. This in part appears to reflect a level of trust in humanitarians over the government:

[T]here’s no danger in it, there’s humanitarian, there’s no danger unless it was the government they will do something different, if there’s a humanitarian there’s no threat or problem. (Female IDP, SS.02.IDP.04)

[W]e assumed that since we left our farm we came to PoC, we assumed that UN will not harm us then the information we are giving will only be used in a good way so we didn’t … ask, we only know this information is used in providing services to us. (Female IDP, SS.02.IDP.02)

The magnitude of IDP needs and the limited resources of local agencies was also highlighted by some local stakeholders as a factor limiting the potential for a focus on data literacies and empowerment:

International organisations with funding are able to collect data in a very professional way and using up-to-date gadgets … While the data is crucial to humanitarian assistance, funding allocated to it is only possible for the big UN agencies and international organisations … we know data is important, but we do not have funding for it. (NGO Representative, SS.02.SH.10)

Engaging IDPs and enhancing data literacy can be difficult, resource intensive and time-consuming work. The levels of sustained engagement to leverage the transformative potential of such processes are limited in some contexts and can even appear unnecessary due to generalised acceptance of data collection on the part of IDPs. Nevertheless, our analysis also suggests that investment in such processes is a vital step toward challenging entrenched asymmetries within the humanitarian field.

While initial engagement with our visual materials was limited in South Sudan, we also found that over time IDPs began to express interest and wanted to be more involved as data actors – even in contexts of extreme need, and despite being poorly equipped to understand issues surrounding data collection and use. For example, in South Sudan, IDPs who attended our feedback workshops began to engage with questions around data rights and justice. While many initially appeared to be disinterested in the use of their data, in some cases this changed throughout the research process and the processes of mutual learning that our ongoing engagement initiated. Trust in our project also began to build as this understanding developed. This is evident, for example, in the statement of one of the participants of our Juba workshop:

I am so glad to be here today in this colourful event. This means a lot to us. When you came to interview us, I was so scared that you may take our data and get us arrested by the government and that is because at the time of the interviews, you were recording us. Now I and the community members trust you completely. You are now free to come back and we will not be afraid. I will personally allow you to interview my wives, my children and my parents. (Male IDP, Juba Workshop 24.8.22)

This quote also illustrates the importance of recognising the heterogeneity experiences under the label of ‘IDP’. The gender dynamics are most obvious here in the statement that ‘I will personally allow you to interview my wives’. However, we are also reminded here of a wider set of dynamics within affected communities which mediate the who and the how of participation in research and data collection processes. This may relate to gender but also age, ethnicity or socio-economic position.

Despite the very real challenges of critical data literacy in contexts such as South Sudan, our project suggests that such efforts can have transformative effects. Although not always feasible, we nevertheless want to highlight how an ongoing commitment to working through data with affected communities over time may be critical in challenging the extractive dimensions of data-driven humanitarianism and the asymmetric relations that this reinforces (Squire and Alozie, 2023). A discussion of our work in the Nigerian context provides further insights here.

Collaboration and Co-production

In Nigeria, the Data and Displacement team worked beyond the initial scope of the project to find ways to include IDPs and local stakeholders more substantively throughout the project cycle. As indicated above, we first undertook interviews with fully informed consent, using the simple infographic of the data journey in order to sensitise research participants to our project themes and to begin to unpack understandings around data and data flows. We found this to be quite challenging in the first instance and worked hard to support the researchers who were undertaking interviews to unpack the infographic with research participants. We then engaged selected IDPs (three women and four men), as well as local stakeholders, in participatory workshops to facilitate review of the complex infographic at our dissemination event in Abuja in July 2022. This enabled us to glean further insights into areas of friction surrounding the movement of data at different stages of the journey.

The participatory workshops were driven by our critical pedagogical approach. Instead of engaging with the complete picture of the complex data journey, we initially opened discussion about the main data practices involved: data collection, data management, data and data usage. We then worked with participants to ground these visuals in their situationality (see also Porto de Albuquerque et al., 2023), that is, to discuss their lived experiences along the data journey by asking situated questions such as: Which types of data collection have you experienced? Which organisations have collected data from you? Do you know what they do with these data? Do you know how they used these data later? Did you see any results coming from the data you provided? As IDPs discussed these questions, we wrote keywords of their answers in sticky notes and glued in the corresponding places in the visual diagram (Figure 1). This process sparked reflection and discussion among the participating IDPs, who discussed and mutually enriched their understanding of the journeys followed by their own data produced by humanitarian organisations.

View Full Size

Grounding the data journey

Citation: Journal of Humanitarian Affairs 5, 3; 10.7227/JHA.113

• Download Figure
• Download figure as PowerPoint slide

Feedback during our workshops indicated that this usage of the data journey visual provided an important pedagogical tool for both practitioners and IDPs alike. Using our data journey visualisation as a starting point, IDPs and local stakeholders were able to reflect on their lived experiences with data collection practices and to discuss their expectations on what happens with their data, including how, by whom and for what purposes the data is used. This began to advance awareness of data processes. For instance, many IDPs were able to identify the contexts and situations of the data journey and the crucial roles they can play in ethical data processes, both in and out of the camps. At the end of the event, our research participants gave feedback on what they learnt relative to ethical data processes in context of the project. They suggested that they better understood the importance of data, had an appreciation of their data rights and that they were better equipped to ask appropriate questions on ethical data processes and enforcement of their rights. As one IDP in Abuja summarised: ‘We have data rights.’

The Abuja event was followed up with another more focused workshop in Lagos in January 2023. This was designed as a co-production workshop to enable IDPs to engage and defend their data rights and communicate these to the world through the production of a digital output for wider use. It involved five of the IDPs (two women and three men) who had previously participated in the Abuja event. Those involved expressed appreciation of the complex infographic as a data-literacy learning tool and took pleasure in identifying their contributions on the poster presentation directly. For example, one pointed to a part of the infographic and said: ‘That’s my idea!’ The workshop sought to engage with IDPs in discussion around data rights and the importance of data ethics. It led to the co-production of a series of data clips in four languages on these issues. Members of the project team deliberated with research participants about what questions should be included to increase awareness within their communities of data rights, questions to ask data collectors, what to expect during data collection processes and the importance of feedback as critical aspects of ethical data collection and management ecosystem for transparent and accountable humanitarianism and displacement interventions (Figure 2). The questions were then scripted collectively, before being recorded by IDPs in Hausa, Kanuri and Pidgin English languages for accessibility. A translation into Nuer for use in South Sudan was produced after the event.

View Full Size

Draft questions for digital rights clip

Citation: Journal of Humanitarian Affairs 5, 3; 10.7227/JHA.113

• Download Figure
• Download figure as PowerPoint slide

The co-production workshop required significant time and resources and was only undertaken with five IDPs. However, their involvement in the scripting and recording of clips in languages of their choice meant they directly provided insights into how members of affected communities can engage stakeholders in relation to data rights and data ethics. The longer-term and wider community impacts of such work remains to be seen, but in a more immediate sense those who participated in the process also developed data-literacy skills through their scripting and recording of short videos for members of their communities. Nevertheless, it is in the broader critical reflection on the communication barriers that are exacerbated by data-driven humanitarianism that the importance of this intervention lies (Squire and Alozie, 2023; Jones and Logo, in progress). This reflects our key aims to collectively and inclusively develop accessible outputs with and for affected communities, in order to reflect on and build capacities in the areas of data ethics and data rights and based on sustainable, regenerative and transformational learning and knowledge impacts on affected and vulnerable groups in context of humanitarian crises. This was developed against the general, and specific, background of project participatory principles that prioritise ethical knowledge and data processes that include feedback and inclusiveness. A focused and sustained engagement with IDPs through the development and co-production of tools for data literacy has been an important factor in this process – but one that can be difficult in some contexts and that requires different levels of effort and capacity to achieve.

Conclusion

This article has shown how using visual resources in contexts of precarity, marginalisation and deprivation proved to be both illuminating and challenging in our research. The IDPs we interviewed often articulated an interest in knowing more about why and how their data is collected, stored and used in humanitarian decision-making, while others appeared disinterested. We also identified complex relationships of authority, fear and need, potentially preventing probing questions from being asked of humanitarian data collectors. In such a context, visual resources such as infographics provided important knowledge and generated space for members of affected communities to ask questions and reflect on the data journey. However, in other instances it continued to be difficult for our research participants to understand and engage with the data visualisations. Our research thus corroborates what others have found: that the generation of knowledge in contexts of violence and division is influenced by, and reinforces, unequal relationships of power (Chonka et al., 2022: 568). What our findings additionally show is that critical data literacy as a transformative tool can only emerge from participatory processes of sense-making that are grounded in the specific situations and lived realities of IDPs.

Context matters, and the time and resources required to engage participatory work with people in contexts of displacement are significant. Even in less demanding contexts, research is too often identified as extractive in nature, with researchers ‘taking’ data from marginalised communities to use in settings of benefit to the researcher, while additional and contextually informed insights are lost (Gaudry, 2011: 113). This calls for a collective commitment to meaningful engagement with affected communities through the co-production and shared reflection on the production and use of data. For example, humanitarians could consider incorporating data-literacy assessments of the IDP communities with whom they work, protecting funding for data ethics training and implementation and ensuring consistent feedback to affected communities about how their data is generated, stored and used. This requires a shift from thinking of the data as a tool of humanitarian work, towards understanding data as co-produced with displaced people. Researchers wanting to employ more participatory methods to enhance the data literacy and rights of affected communities need to be flexible in the way methods are applied and adapted in context and with feedback of displaced people themselves. It is particularly important in such contexts to understand research beyond a one-off ‘encounter’ and as a longer, iterative process of engagement between the researcher and the researched. This is a challenge not only for researchers who design and implement projects, but also raises questions for the research institutions in which researchers are embedded, including their ethics, funding and data procedures which often delineate what the lifetime of a project, and therefore participation in it, should be (Chonka et al., 2022).

Further developing a critical pedagogical approach to data for the humanitarian context, we thus propose that such participatory data practices don’t simply amplify IDP’s voices and make their views better represented in humanitarian data. Notwithstanding the challenges, they can also contribute to the development of critical consciousness about IDP’s lived experiences as humanitarian data subjects. Yet even if we see the potential of this critical approach to challenge unjust and non-inclusive data and humanitarian practices, we are also conscious that there are a range of other factors that will influence the transformation of wider structures of dependency and oppression in the humanitarian sector that need to be addressed. Our project experiences and research findings thus caution against viewing data literacy as an easy fix to empowerment challenges, both in the conduct of humanitarian work and in the implementation of research. Indeed, our starting point is not that IDPs need to learn more about how their data is generated and used in a decontextualised manner, but rather that there is a need for broader critical reflection on the material and communicative barriers that are exacerbated by data-driven humanitarianism. The work of transformation through critical data literacy thus also requires the dismantling of deeper-rooted power asymmetries and relations of dependency within which data-driven humanitarianism is embedded. While an important endeavour, critical data literacy is not an easy fix to empowerment challenges, either in the conduct of humanitarian work or the implementation of research striving to use more participatory methods. Nevertheless, we hope that this article has positive implications for the production of inclusive and ethical data through the development of more effective communications within humanitarian ecosystems and the scholarship informing these.

Acknowledgements

This article draws on research funded by the UK Arts and Humanities Research Council and Foreign, Commonwealth & Development Office (AHRC-FCDO) Collaborative Humanitarian Protection Programme grant AH/T007516/1, Data and Displacement: Assessing the Practical and Ethical Implications of Targeting Humanitarian Protection. The authors would like to extend many thanks to the wider research team for their work on this project, including Olufunke Fayehun, Leben Moro, Dallal Stevens, Rob Trigwell, Modesta Alozie, Prithvi Hirani, Grant Tregonning, Stephanie Whitehead, Abubakar Adam, Hajja Kaka Alhaji Mai, Omolara Popoola, Silvia De Michelis, Ewajesu Opeyemi Okewumi, Mauricio Palma-Gutiérrez, Funke Caroline Williams and Oluwafunto Abimbola. We would like to sincerely thank all our research participants, including those who engaged in our participatory workshops in Nigeria and in our feedback sessions in South Sudan. Thanks are also extended to Hajia Yabawa Kolo, Director General at Borno State Emergency Management Agency (2023), to the project photographer, Mark Faeren Hirekaan, to our two anonymous reviewers and to the journal editors.