Supporting integration in schools: Evidence from a randomized targeted intervention

We evaluate a two year school development program aimed at enhancing the quality of education for recent migrant students and Swedish language learners through customized packages of professional development for teachers and support of school man-agement. We exploit the pairwise randomized roll out to 63 municipalities between 2016 and 2019 to examine effects on student achievement and explore underlying mechanisms. Over a 7 year follow-up period, core subject test scores improved by 0.021 sd, driven by a 0.032 sd improvement in mathematics performance. Test score gains in mathematics were present for students, regardless their background, during and post-implementation. Swedish test score gains generally materialized post-implementation, while second gen-eration immigrant students gained already during implementation. Test score gains are more pronounced for foreign background students, for boys and at the low end of the test score distribution. The support program passes a cost-benefit test.