Science
Volume 359, Issue 6373, 2018, Pages 325-329

Improving refugee integration through data-driven algorithmic assignment (Article) (Open Access)

Bansak K. , Ferwerda J. , Hainmueller J. , Dillon A. , Hangartner D. , Lawrence D. , Weinstein J.*
  • a Department of Political Science, Stanford University, Stanford, CA 94305, United States, Immigration Policy Lab., Stanford University, Stanford, CA 94305, United States, ETH Zurich, Zurich, 8092, Switzerland
  • b Immigration Policy Lab., Stanford University, Stanford, CA 94305, United States, ETH Zurich, Zurich, 8092, Switzerland, Department of Government, Dartmouth College, Hanover, NH 03755, United States
  • c Department of Political Science, Stanford University, Stanford, CA 94305, United States, Immigration Policy Lab., Stanford University, Stanford, CA 94305, United States, ETH Zurich, Zurich, 8092, Switzerland, Graduate School of Business, Stanford University, Stanford, CA 94305, United States
  • d Immigration Policy Lab., Stanford University, Stanford, CA 94305, United States, ETH Zurich, Zurich, 8092, Switzerland
  • e Immigration Policy Lab., Stanford University, Stanford, CA 94305, United States, ETH Zurich, Zurich, 8092, Switzerland, Center for Comparative and International Studies, ETH Zurich, Zurich, 8092, Switzerland, Department of Government, London School of Economics and Political Science, London, WC2A 2AE, United Kingdom
  • f Immigration Policy Lab., Stanford University, Stanford, CA 94305, United States, ETH Zurich, Zurich, 8092, Switzerland
  • g Department of Political Science, Stanford University, Stanford, CA 94305, United States, ETH Zurich, Zurich, 8092, Switzerland

Abstract

Developed democracies are settling an increased number of refugees, many of whom face challenges integrating into host societies. We developed a flexible data-driven algorithm that assigns refugees across resettlement locations to improve integration outcomes. The algorithm uses a combination of supervised machine learning and optimal matching to discover and leverage synergies between refugee characteristics and resettlement sites. The algorithm was tested on historical registry data from two countries with different assignment regimes and refugee populations, the United States and Switzerland. Our approach led to gains of roughly 40 to 70%, on average, in refugees’ employment outcomes relative to current assignment practices. This approach can provide governments with a practical and cost-efficient policy tool that can be immediately implemented within existing institutional structures. © The Authors.

Author Keywords

[No Keywords available]

Index Keywords

refugee human community integration Refugees priority journal data synthesis machine learning supervised learning algorithm Algorithms United States Humans integrated approach population research Article migration data analysis Emigration and Immigration employment Switzerland process development
Link
https://www.scopus.com/inward/record.uri?eid=2-s2.0-85040790826&doi=10.1126%2fscience.aao4408&partnerID=40&md5=eba3a44aef5c01084a6e903d0e903ce2
DOI: 10.1126/science.aao4408
ISSN: 00368075
Cited by: 20
Original Language: English