Over the last decade, efforts to build the evidence base for humanitarian action have grown immensely. While localisation is often discussed in relation to programming, the same approach is often missing in research and evaluation. Researchers based in the Global North often work with teams of enumerators and facilitators from the countries under study to collect data; however, the analysis, publication and dissemination of these findings is often carried out away from these settings and without the participation of local humanitarian workers, let alone members of the affected community. The way that imbalances of power affect the health and well-being of crisis-affected populations is an increasingly common topic of study, and considering power within research processes is equally important.
In 2019, we partnered as a national NGO based in Lebanon (Union of Relief and Development Associations (URDA)), an international NGO (CARE in Lebanon) and a Global North-based research institute (the Global Women’s Institute at the George Washington University (GWI)) to carry out participatory action research with women and girls living as refugees. The project, Empowered Aid, is based at GWI and takes place in Lebanon and Uganda – two of the largest refugee-hosting countries in the world. It was initially designed with a number of partners in both countries, some of whom had established working relationships with GWI staff. The Institute has recognised expertise in ethical and participatory research on violence against women and girls in humanitarian and development settings+1. See for example: GWI (2017) Bridging research and action for change: 5-year anniversary report. The Global Women’s Institute at the George Washington University (https://globalwomensinstitute.gwu.edu/sites/g/files/zaxdzs1356/f/downloads/GWI_5yearReport_FINAL_Proof_lo-res_spreads.pdf).. The research phase in Lebanon was designed through workshops held with CARE and URDA. Women and girls are centred as knowledge-holders who identify and share the ways in which accessing humanitarian aid – such as cash assistance, food, fuel and firewood, shelter and water, sanitation and hygiene (WASH) – can put them and their peers at risk of sexual exploitation and abuse (SEA). Recommendations were also developed to better address such risks and these were applied to distributions, where safety and risk were measured using adapted distribution monitoring tools.
The findings of our research and all the participatory methods and tools we have developed and used are freely available+2. To read more about Empowered Aid and access our results reports, tools, facilitation guides and manuals, visit https://globalwomensinstitute.gwu.edu/empowered-aid. Peer-reviewed publications of the research findings are forthcoming.. In this article, we reflect on our research process in Lebanon: from design and planning, to data collection and analysis, to sharing findings and putting them into action. At each stage, we highlight aspects of our partnership that we consider innovative; how we sought to take a capacity-sharing approach at each stage of the process; and how risk was managed and shared, particularly since the onset of the Covid-19 pandemic.
Research design and planning
In selecting a distribution partner with whom to test Empowered Aid’s recommendations, several criteria were considered important. First, that the partner was engaged in distribution programming in northern Lebanon that included at least one of the types of aid mentioned above. Next, that they had incorporated gender and protection mainstreaming into their distribution programme, as well as a mechanism for safeguarding. And finally, that they were open and willing to commit to a research and learning process that centres women’s and girls’ experiences and includes their voices in humanitarian decision-making. URDA was interested in this partnership in order to share their deep knowledge of the communities in which they work, and to benefit from learning and exchange with organisations focused on gender and protection issues. An initial design workshop was held in person (before the pandemic) where each partner presented their organisation’s mission and structure, reviewed findings from the first phase of Empowered Aid and identified ways to apply them in URDA’s distribution programming. In these workshops, the partners also shared their distribution monitoring tools and jointly adapted them to better measure SEA risks, based on findings from the research+3. This article summarises the field-testing phase for these tools, which is now concluding. The tools will soon be available as part of a comprehensive toolkit that will be shared on the website (https://globalwomensinstitute.gwu.edu/empowered-aid)..
With the onset of Covid-19, the team had to find new ways of working. Even CARE and URDA staff based in the same country could not meet in person for long periods. In order to ensure all partners could continue to contribute to project design and management discussions, extra mobile internet was purchased for team members from the national NGO, as they tended to have less reliable access to internet, and the international team collaborated via Zoom for online workshops scheduled to fit the Lebanon timezone. SEA awareness messages were updated and accompanied by Covid-19 awareness messages in all distribution and monitoring activities.
It was critical that major decisions be made in collaboration, with each partner organisation having an equal voice in the process. In this way, the team weighed benefits of continuing the research against possible risks (a key ethical principle of all research), given the changing context of the pandemic. As lifesaving aid was being distributed, it was decided that the distributions we had planned and procured would go ahead even though much of the data collection, such as focus groups and in-person surveys, would need to be dropped due to public health regulations. Instead, a new tool was developed, a short 4–6-question in-person interview that could be conducted quickly at the distribution site. This allowed for basic information on safety and access to services to be collected from women and girls at a time when service providers, such as CARE and URDA, were largely cut off from these populations, whose lower access to mobile phones made them less able to engage in the shift to remote services under Covid-19.
Data collection and analysis
With the onset of the pandemic, risk management came to the fore as URDA and CARE staff (following their agencies’ protocols) decided whether to engage directly in the distributions. It was a difficult decision: URDA was aware that most NGO staff were at home while refugee communities were in even more need given the combination of lockdown and Lebanon’s economic crisis. The principle of ‘do no harm’ took on new meaning as the URDA, CARE and GWI team worked together to avoid exposing people to additional risks. As knowledge about the virus changed rapidly and public health guidance shifted during the first part of the pandemic, we changed distribution plans six times. Ease of communication was essential and URDA’s practice of using Whatsapp inspired the research team to create a WhatsApp group to share challenges on the spot and collectively identify ways to overcome them. Every night at least some members of the team held meetings to discuss and re-plan: even sharing a pen (for signatures used in verification processes) could potentially cause harm. Continual ‘reinventing’ was exhausting, but was key to maintaining safety for ourselves and those we were serving.
Key elements of the adapted distributions included ensuring that teams were provided with adequate protective equipment; URDA staff still reflect on how much they miss sharing smiles with community members receiving the aid distributed. Distribution planning was spaced and timed to reduce the number of people at the site at any one time. Coincidentally, many of these modifications aligned with women’s and girls’ recommendations for reducing SEA, which we were already planning to incorporate, such as having a smaller number of families arrive at pre-arranged timeslots to collect aid rather than gathering large numbers together as is done typically. For example, in a tented settlement of 315 families in Akkar, we divided the families into ten groups with approximately 30 people visiting the distribution site at a time. Chairs were set out at two metres apart to ensure adequate spacing and aid recipients were called one by one to receive their food parcel and return home. Chairs were sanitised before the next group came. At another distribution site, aid was distributed door to door by mixed-sex teams – another recommendation from Empowered Aid’s first phase, as women and girls reported feeling unsafe when men aid workers visited their homes alone.
In both of these modes of distribution, awareness sessions were conducted on SEA risk and Covid-19 key messages. Distributions were monitored using safety audits and a short in-person questionnaire (described above) that was asked of a subset of adult women attending the distribution. In-person post-distribution monitoring (PDM) surveys were shifted to phone surveys and carried out with a subset of all distribution participants several weeks later, to assess their satisfaction and perceived safety with these new methods of distributing. Refugee families generally reported more concern about lack of food than Covid-19: ‘If we die from Coronavirus and illness, it will be better for us than to die from hunger!’.
Sharing findings and putting them into action
During this participatory action research, which is ongoing, Lebanon has faced revolution, financial crisis, a global pandemic and, on 4 August 2020, one of the largest non-nuclear explosions ever recorded when 2,750 tonnes of improperly stored ammonium nitrate exploded in Beirut’s port. URDA and CARE, as well as other aid actors in Lebanon engaged as members of Empowered Aid’s technical advisory group, immediately applied the findings of the research to their emergency response efforts in the aftermath of the blast. For example, modes of delivering aid and repair assistance at household level incorporated the recommendation to use mixed-sex or all-women teams, and overall more women aid workers, volunteers and community leadership structures were incorporated into URDA and CARE’s emergency responses. Ensuring that low-literacy, visual awareness materials were posted at key locations such as Western Union offices and cash machines is not as easy as it sounds, especially when teams were already stretched.
The Covid-19 pandemic had already fundamentally changed the way information about services and aid is shared with refugee communities. In many ways we have failed women and girls in our reliance on mobile and online platforms, as many fall into a gendered ‘digital divide’ because dominant gender roles limit their access to any phone, or only to male relatives’ phones. When conducting the phone survey as part of distribution monitoring, for instance, we found we had to call households in the afternoon when men had returned home if we wanted to reach women, so their husbands could pass the phone to them. Our consent process included questions to determine if the person responding had enough privacy to speak fully; when this was not the case, potentially sensitive questions about safety and risk were not asked. At the same time, the constraints on women’s and girls’ ability to access information and services puts them at potentially higher risk of exploitation and abuse by those who may take advantage of this power differential. With this in mind, Covid-19-specific recommendations were drawn from Empowered Aid’s research+4. Empowered Aid’s Covid-19-specific findings and recommendations for Lebanon are available at: https://globalwomensinstitute.gwu.edu/sites/g/files/zaxdzs1356/f/downloads/GWI-CARE-Empowered%20Aid-Lebanon-Covid19Note_a11y.pdf. Sector-specific recommendations are available at: https://globalwomensinstitute.gwu.edu/empowered-aid-resources..
Throughout the research, we found it essential to not import foreign or pre-determined processes for data collection, analysis and write-up, but rather to co-create processes based on local partners’ knowledge of the rapidly changing Lebanese context and programming realities. This became even more essential with the onset of the pandemic when our internationally based team members (some of whom had spent years working in Lebanon) could no longer travel. Collaboration between URDA, GWI and CARE strengthened the bond between the organisations as well as individual team members, who felt that they were able to maintain a voice in decision-making and common vision, despite working in such a dynamic environment. We feel the learning generated by Empowered Aid was produced and shared in ways that prioritise local actors’ needs and encouraged creative and innovative thinking at a truly extraordinary time.
Jihan Kaisi is the Executive Director of URDA (firstname.lastname@example.org). Rosy Haddad is a Board Member and Public Relations Officer for URDA (email@example.com). Loujine Fattal is the Empowered Aid Research Manager with CARE in Lebanon. Alina Potts is a Research Scientist with GWI at the George Washington University.