The Impact of Culture on Migrant Integration: Socioeconomic Meta-Analysis of the Epidemiological Literature [CULTIMA]

我们请求资助一项关于文化背景对移民行为影响的元分析。移民多快和在多大程度上融入并可能被东道国同化的问题具有高度的政治、社会和经济重要性。因此，实证研究迅速发展，经济学和社会学的研究并行，但大多是脱节的。文献主体正变得越来越难以总结，原因如下：首先，研究结果多种多样，总体上尚无定论。其次，研究在研究设计方面存在很大差异，例如，在数据、样本选择和建模方面。第三，选择性移民和一体化的制度性障碍，如歧视，是阻碍确定因果关系的经验挑战。我们提出的荟萃分析提供了一个全面和系统的文献综合，推动了这个固有的跨学科主题的单学科研究的局限性。元分析工具将帮助我们评估和解释不确定的发现。因此，我们可以在研究来源国对移民行为的影响时，测试并在必要时纠正发表偏倚（只有统计上显著的预期结果才被发表的可能性）。元回归将允许我们探索方法的异质性，并得出比原始研究能够提供的关于祖先文化和个体行为之间关系的更普遍的结论。最后，我们的项目有助于评估由于经验挑战而在测量因果效应时可能存在的偏差的程度。

We request funding for a meta-analysis on the effects of cultural background on immigrant behavior. The question of how quickly and to what extent migrants integrate and, possibly, assimilate to the host country is of high political, social, and economic importance. Accordingly, empirical investigations have been developing rapidly, with parallel but mostly dis-connected research in economics and sociology. The body of literature is becoming increasingly hard to summarize, for several reasons: First, results are diverse and overall inconclusive. Second, studies differ substantially in research design, e.g., with respect to data, sample selection, and modelling. Third, selective migration and institutional barriers to integration, such as discrimination, are empirical challenges that hinder the identification of causal effects.Our proposed meta-analysis provides a comprehensive and systematic synthesis of the literature, pushing the limits of mono-disciplinary research on this inherently interdiscipli-nary topic. Meta-analytic tools will help us to assess and interpret inconclusive findings. Thereby we can test and, if necessary, correct for publication bias (the possibility that only statistically significant results of the expected sign are published) in research on source country effects on immigrant behavior. A meta-regression will allow us to explore methodological heterogeneity and draw more universal conclusions on the relationship between ancestry culture and individual behavior than primary studies are able to provide. Finally, our project contributes to assessing the extent of a possible bias in measuring causal effects due to empirical challenges.