:focal(3629x1117:3630x1118))
Using machine learning to help refugees succeed
How GeoMatch is revolutionizing resettlement efforts.
)
Dominik Rothenhäusler grew up in Oberzell, Germany, a town of roughly 2,500 people along the Schussen River. Like many towns and cities across Germany, Oberzell has absorbed a surge of asylum seekers and refugees in recent years—at first, mostly people from Gambia, Senegal, Cameroon, and Afghanistan; more recently, families and individuals feeling the war in Ukraine have entered Oberzell in need of asylum.
Rothenhäusler, now an assistant professor of statistics in the School of Humanities and Sciences at Stanford University, mostly watched from afar as his hometown endeavored to support the refugees. His old soccer coach emerged from retirement to host weekly practices and scrimmages. Other residents stepped in to show the ropes of riding public transportation or navigating municipal bureaucracy. Volunteers taught basic German.
“I was somewhat separated from all of it, but I wanted to do my part,” Rothenhäusler says.
It was convenient, then, when the Stanford Immigration Policy Lab (IPL) reached out asking for help on a refugee placement project called GeoMatch.
“With academic research, there are usually a few steps between the work and its impact. With this project, there was a clear pathway to immediately effecting positive change in the world,” he says.
)
A train in Switzerland. Photo: Yann Lerjen, Unsplash
Better living through machine learning
GeoMatch is a machine learning tool designed to help placement officers match refugees with the communities where they’re most likely to thrive. It’s partially funded by a Stanford HAI Hoffman-Yee Grant, which are designed to address significant scientific, technical, or societal challenges requiring an interdisciplinary team and a bold approach. Stanford Impact Labs is also supporting the project with funding to test the tool’s impact with nonprofit partners.
By applying an algorithm to assign refugees to a location where they would be most likely to succeed, the research team was able to increase projected employment by roughly 40 percent in the United States and 75 percent in Switzerland. The researchers published their results in Science.
The idea for GeoMatch originated when a team of researchers—including Jens Hainmueller, co-director of the Stanford Immigration Policy Lab and the Kimberly Glenn Professor and professor of political science in the School of Humanities and Sciences—met with U.S. government and nonprofit agencies that assist with refugee placement and integration.
At the meeting, conversation turned toward the challenges faced by placement officers. Though the resettlement process raises many questions—when are cities better for refugees and when are rural areas? Are homogeneous or diverse communities preferable? What local resources contribute to job placement?—none had been formally investigated. Instead, placement officers relied primarily on experience and intuition when finding new homes for refugees.
“So much data exists in these management and administrative systems, but historically it has been challenging to use it effectively,” says Michael Hotard, the director of GeoMatch. “We started asking how the information could be harnessed to help the people making these placement decisions.”
Hainmueller and several colleagues got to work creating an algorithm that centered on refugee placement in the United States and Switzerland. The algorithm matched a range of individual background characteristics—country of origin, language skills, gender, age—with a refugee’s time of arrival and assigned location; for outcome, the algorithm measured employment success 90 days after arrival in the United States and three years after arrival in Switzerland. (These benchmarks of success are used by the U.S. and Swiss governments, respectively.)
At the heart of this algorithm-based project is, of course, a deeply human reality. More than 100 million people around the world have been forced to flee their homes, according to the United Nations. About 35 million of these currently displaced people are recognized as refugees, nearly half of whom are under the age of 18. Given such numbers, the algorithm’s effectiveness raised the possibility of dramatically improving millions of lives. Hainmueller and his colleagues wanted to lift this work from the pages of academic journals and get it into the hands of those working daily on the issue of resettlement.
How it works: GeoMatch’s algorithm studies data from past immigrants such as work history, education, and personal characteristics to find patterns about what made people more or less likely to succeed. Then, it predicts a new immigrant’s likelihood of success at a range of locations within the destination country and recommends them to the newcomer.
The road to collaboration
The large refugee resettlement nonprofits in the United States receive the bulk of their funding from two federal agencies: the State Department and the Department of Health and Human Services. Shortly after the scholars published their initial paper, additional organizations reached out to IPL to discuss how GeoMatch might help their organizations.
Since 2020, Switzerland has been testing GeoMatch in its placement of refugees around the country. “We wanted to build rigorous impact evaluation into this program to make sure that the tool we’re developing achieves the effects we would expect,” Hotard says. The program has been rolled out as a large-scale randomized controlled trial.
Asylum seekers entering Switzerland have traditionally been assigned to one of the country’s 26 cantons, or administrative states, based on the need to balance population distribution rather than a desire to find the best economic fit. The Swiss Secretariat for Migration is piloting GeoMatch to help with location decisions for a portion of incoming asylum seekers. Among the asylum seekers in the pilot, roughly half will receive a GeoMatch recommendation designed to maximize employment prospects after three years. The rest will follow the traditional route of assignment.
In addition to helping refugees more quickly integrate into new economies, this process dramatically lightens the administrative burden placed on host countries. GeoMatch can incorporate the various constraints that placement officers face when trying to find the right location, such as medical needs, family size, languages spoken, and schooling needs.
“What once took multiple people hours of research can now be done in minutes,” Hotard says. “GeoMatch can be incredibly useful as a tool that simplifies the process of gathering information and making connections. It automates much of what has traditionally been done manually.”
“What once took multiple people hours of research can now be done in minutes. GeoMatch can be incredibly useful as a tool that simplifies the process of gathering information and making connections. It automates much of what has traditionally been done manually.”Michael Hotard, director of GeoMatch
)
Signs in a Geneva train station. Photo: Cybrneon, Unsplash
Building guardrails and new applications
Automation, of course, raises concerns. The refugee resettlement process depends on highly sensitive information, and the results of each placement decision are profoundly consequential. For this reason, placement officers will serve as the final decision makers when considering where refugees land.
“We’ve been hearing more and more from our partners that they’re thinking through issues of fairness in the algorithm,” says Elisabeth Paulson, a professor at Harvard who recently completed a postdoc at IPL. “My work at IPL tried to preemptively ensure these algorithms produce fair outcomes.”
The aim of these efforts is a complementary relationship between technology and the humans using it. With several colleagues, Paulson crafted a modified set of algorithms that allow resettlement agencies to carefully examine results like employment across distinct subgroups. If one agency in the United States wants to make sure specific employment thresholds are achieved based on a refugee’s country of origin, for instance, this can be set as the desired outcome. Another agency could instead tune the algorithm to look at employment rates based on gender. The goal is to maximize overall employment without accidentally harming one subgroup.
In a related push to improve the algorithm, Rothenhäusler was recruited to the GeoMatch team because of his expertise in “distribution shifts.” In essence, the fact that machine learning models are trained on historical data creates unseen problems when present-day circumstances don’t mirror historical circumstances. Rothenhäusler noted that looking for employment today in the Bay Area is a different endeavor than it was two years ago; likewise, asylum seekers showing up in Europe eight years ago—many from Africa and the Middle East—are different in important ways from the Ukrainians seeking asylum there today. Rothenhäusler’s work builds resilience against these kinds of changes into GeoMatch.
In the future, the group hopes to increase the number of countries it partners with—work is ongoing in the Netherlands—as well as the populations that it serves. An emerging collaboration in Canada, for instance, is testing how GeoMatch might assist economic immigrants rather than refugees, matching their individual skill sets and location preferences with the best localities for them and their families. A domestic partnership with Lutheran Immigration and Refugee Service has also launched.
The number of refugees and migrants now is unprecedented; as war, economic hardship and resource scarcity persist in the years ahead, the amount of suffering will too. Hainmueller says the goal is to continue expanding the use of GeoMatch, to improve outcomes throughout the world.
“The countries we’re working with today are just the beginning,” he says.
:focal(1071x1527:1072x1528)){kind=spacer}
:focal(3159x1283:3160x1284))
:focal(1099x1103:1100x1104))