Filling the Forced Displacement Evidence Gap: Taking Stock on World Refugee Day

Filling the Forced Displacement Evidence Gap: Taking Stock on World Refugee Day

Template G Content Blocks
Sub Editor

By Nessa Kenny and Laura Benrey

Cox's Bazar
Photo: Sebastian Chaskel


Over 70 million people are forcibly displaced as a result of conflict, persecution, or violence—more than ever before in history. Displaced communities, who were already vulnerable before the pandemic hit, are particularly at risk now: they fall outside of traditional social safety nets, many live off of earnings made or food purchased that day, and many live in densely populated refugee camps where social distancing and handwashing are difficult, if not impossible. Last month, for example, our Cox’s Bazar Panel Survey (a representative panel of Rohingya refugees and host community members) found that 70 percent of Rohingya refugees have been unable to buy essential food in the past week due to the impacts of COVID-19.

Aid organizations responding to the pandemic are operating with little evidence on how to invest their limited resources, even in ordinary times. Last year, we started reviewing the existing evidence on what works to support and improve outcomes for forcibly displaced populations. The body of research is limited—but the good news is, this is changing quickly. 

Many Challenges, Little Evidence 

While there is a growing empirical literature on forced displacement, there are fewer than 25 published randomized evaluations that evaluate programs for displaced populations. Randomized evaluations are far from the only form of rigorous evidence, but this number serves as one indicator of the paucity of evidence on this topic and provides a window into how certain methods can be difficult to apply to issues of forced displacement. 

Aid organizations responding to the pandemic are operating with little evidence on how to invest their limited resources, even in ordinary times.

There are plenty of good reasons why this number is so low: it is very hard to do impact evaluations in crisis-affected contexts. Instability, insecurity, and new displacement influxes force organizations to adapt quickly, causing programs to rapidly move from design to implementation, to adaptation. Population mobility means that displaced individuals may not complete programs they are enrolled in, or may be very hard to follow up with. Implementers often have questions about the ethics of randomization or research with refugees and internally displaced people. Administrative data is rare or unreliable. In most places with large displacement influxes, there isn’t strong research infrastructure, increasing study costs. The list could go on, but all of these factors make it difficult to design and implement impact evaluations on forced displacement. 

The little evidence that does exist overwhelmingly focuses on refugee and IDP mental health; almost 60% of completed evaluations are of psychosocial support programs, with other studies focusing on other types of service provision and social integration. This evidence is undoubtedly important, and it’s contributing to programming in important ways, but much more, on a broader range of topics, is needed.

A Changing Research Landscape

The past three years have seen a significant uptick in forced displacement research. To date, we have identified almost 40 ongoing randomized evaluations of forced displacement programming, meaning there are more randomized evaluations ongoing now than have ever been published on the topic. 

IPA is involved in over half of these studies, and we have plans to expand this portfolio significantly in the coming years. In December, we launched a Humanitarian and Forced Displacement Initiative to do just that. Through the initiative, IPA will continue to implement and fund some of the first rigorous evaluations and other research related to humanitarian crises and forced displacement. We are also overcoming many of the challenges to conducting rigorous research in humanitarian settings by building research and policy infrastructure in several locations, starting with Bangladesh, Tanzania, and Colombia. This infrastructure will allow us to collect high-quality data, run rigorous evaluations, and develop the needed long-term relationships with implementers and local stakeholders to ensure support for the production and use of evidence. You can read more about this work, and the studies we are implementing, here


As the number of studies grows, the focus of this work is diversifying. While previous impact evaluations largely focused on the mental trauma associated with fleeing conflict, ongoing work asks a broader range of questions. We see four emerging themes in this new wave of research.

How should services, such as education and health, be delivered to displaced populations? (45 percent of ongoing studies) 

Service delivery is especially complex for populations on the move and in settings such as refugee camps and informal settlements. In Uganda, for example, a research team is examining the impacts of the Graduation approach—which provides holistic livelihood support to poor households—in a refugee settlement and nearby host communities.  

How do people make decisions related to displacement? (8 percent of ongoing studies) 

When people are forced to leave their homes involuntarily, there are still many choices to make: how far to go and where to settle; whether to do so within formal systems or informally; whether to remain in their host country, wait to be resettled, or repatriate. In Mozambique, for example, a research team is examining whether displacement due to Cyclone Idai impacts health decision-making and whether a risk coping program implemented before displacement can mitigate those effects.

What interventions can improve social cohesion? (26 percent of ongoing studies)

The arrival of displaced populations can lead to tension—ranging from mere frustration to outright violence—between host communities and displaced populations. How can these risks be mitigated and groups be integrated? For instance, researchers are examining how soccer leagues that pair refugees with native-born youth in Lebanon impact social cohesion, building on promising evidence from Iraq.  

How can displaced people be integrated into host labor markets? (21 percent of ongoing studies) 

As the average duration of crises increases, the ability to work is important for displaced populations to retain a sense of agency, earn a living, maintain or build skills, make productive contributions to their families’ wellbeing, and contribute economically. In Uganda, for example, a study is evaluating whether offering firms subsidies to hire refugees can help reduce discrimination in the labor market.

COVID-19 as a Catalyst 

When the pandemic started, aid organizations struggled to continue providing basic services while also meeting the growing health, employment, and safety needs of already marginalized displaced communities. At this time when our partners needed actionable research and data to guide decision-making, we expected a pause in our forced displacement projects, given the pandemic further complicated an already difficult research environment. 


While some projects did have to pause, we also saw a surprising and heartening expansion of the ongoing work, as researchers moved quickly to answer pressing COVID-related questions for forcibly displaced populations and provide partners with much-needed information. In the last two months, researchers have started or pivoted projects to understand how messaging is diffused in humanitarian settings in Turkey, how cash can help Venezuelan migrants in Colombia, the impact of the pandemic on decisions to return from Lebanon to Syria, and whether graduation mitigates the effect of COVID-19 in Uganda

While some projects did have to pause, we also saw a surprising and heartening expansion of the ongoing work.

With this recent expansion of forced displacement research, including throughout the pandemic, we expect to have more evidence to share in the coming months and years—and will be working hard to share and synthesize these new findings for the decision-makers who need them most as they become available. 


Last year, we started an evidence synthesis to take stock of what rigorous impact evaluations say about what works to support and improve outcomes for forcibly displaced populations. Our working lists of published and ongoing impact evaluations on forced displacement can be found here. We want our synthesis to be as comprehensive as possible, so please get in touch if we’re missing anything by emailing displacement@poverty-action.org!

June 18, 2020