Media Effects on Attitudes towards Immigration. A Multilevel, Multimethod, Longitudinal Comparative Study

经过几十年的内外移民，欧洲社会在种族和文化上比当代历史上任何时候都更加多样化。这一趋势受到公开和政治上的争议，有可能加剧政治两极分化和激进化，从而威胁自由民主国家。许多人缺乏可感知的移民亲身经历。对于大多数公民来说，间接经历--主要是通过媒体传递的--是形成对这一问题态度的最相关来源。因此，媒体在舆论塑造中扮演着重要的角色。最近关于这一主题的研究证明，媒体对移民的信念和态度存在影响。然而，在理论和经验上仍有很大的改进空间。以群体间威胁理论和属性议程设置方法为基础，我们将对媒体在塑造欧洲社会对移民的信念和态度方面的影响进行方法论上的合理调查。我们通过探索媒体报道的变化如何与国民关于移民的不同方面的信念有关，以及这些信念如何影响对不同形式的移民的态度来做到这一点。为了实现这一目标，我们采用了多方法(3种方法)和多层次(2个层次，个人和国家)的设计，从纵向(18年)和国际比较(6个国家)的角度分析了媒体效应。我们将使用两个公开的数据来源：a)来自欧洲社会调查(ESS)的调查数据)和b)来自欧盟统计局、Genesis和欧洲晴雨表的外部综合数据。此外，我们将通过自动分析报纸文章的内容来收集媒体内容数据。因此，我们将使用神经机器翻译(NMT)工具将德语、西班牙语和法语发表的文章翻译成英语，并在所有国家的所有报纸上应用英语代码簿。将使用有监督的机器学习(SML)技术执行属性分析。这一结果将加深我们对媒体报道在一个有争议的社会领域的影响的理解。此外，这项研究还将在不同的时间和国家条件下，对宏观层面的媒体对个人的影响进行实证检验，假设这些情况会影响媒体报道和个人的信仰和态度。从理论的角度来看，这些发现将有助于完善所提出的语境化媒体效果模型。

After decades of internal and external migration, European societies are more ethnically and culturally diverse than ever before in contemporary history. This trend is publicly and politically disputed and has the potential to intensify political polarization and radicalization and thereby threaten liberal democracies. Many people lack appreciable personal experience with migration. For most citizens indirect experiences – mainly transferred by the media – are the most relevant source for attitude formation on the issue. Therefore, the media plays an important role in shaping the public opinion. Recent studies on the subject prove the existence of media influences on beliefs and attitudes towards migration. However, there is still plenty of room for theoretical and empirical improvements. Building on intergroup threat theory and the attribute agenda setting approach, we will provide a methodologically sound investigation on the influence of media in shaping beliefs about and attitudes on immigration in European societies. We do that by exploring how variations in media coverage relate to different dimensions of beliefs about immigration by nationals and how these beliefs influence the attitudes towards different forms of immigration. To achieve this goal, we analyze media effects in a longitudinal (18 years) and international comparative perspective (6 countries), applying a multimethod (3 methods) and multilevel (2 levels, individual and country) design. We will use two publicly available data sources: a) survey data from the European Social Survey (ESS)) and b) exogenous aggregate data from Eurostat, Genesis, and Eurobarometer. Furthermore, we will c) gather media content data by automated content analyses of newspaper articles. Therefore, we will translate articles published in German, Spanish, and French to English by neural machine-translation (NMT) tool, and apply an English codebook on all newspapers from all countries. Supervised machine learning (SML) technique to perform attribute analysis will be used. The results will add to our understanding of the influence of media coverage in a contested social sphere. Moreover, the research will provide an empirical test of macro-level media effects on individuals in the context of varying temporal and national conditions that are hypothesized to influence both the media coverage and individual beliefs and attitudes. From a theoretical perspective, the findings will help to refine the proposed model of contextualized media effects.