Modeling dyadic decision-making processes of regional mobility and their labour market outcomes

本研究项目的目的是对区域流动及其劳动力市场结果的决策过程进行建模。出发点是假设人员流动的决定不是孤立地作出的，而是一个讨价还价过程的结果，这一过程根植于各自的区域和伙伴关系背景中。因此，迁移决策从具有多个解释层次和多个决策阶段的二元视角建模。通过以多层次结构的多元非线性probit模型的形式正式实施空间迁移决策过程，可以对外生迁移决定因素的多层次特征（个人、家庭和环境层面）区分的二元决策过程进行建模。因此，在一个模型中可以同时估计出以往研究中发现的空间流动性决策的多种特征。由于行动者特定迁移倾向和共同迁移决策之间的区别，我们能够考虑二元相互作用以及包括行动者的相对决策权重。因此，有可能将导致决策本身的社会过程纳入其中。此外，我们能够通过评估情境效应和个体-情境相互作用对行动者特定流动性倾向的影响，将当地机会结构的复杂机制纳入多个行动者的决策过程中。在研究项目的第二部分，我们考察了区域流动的劳动力市场结果。在这里，我们能够模拟异质性移动代理的潜在变化的流动性后果，因为劳动力市场资源中特定行为者的差异和特定行为者区域本地化的考虑之间存在明确的区别。因此，有可能确定关键的个体和上下文星座，它们决定了区域移动的利润水平，从而定位经济上多样化的成功移动代理群体。社会经济小组（SOEP）为使用小规模结构特征（SOEP地理数据）的经验实施提供了基础。

Purpose of this research project is the modelling of decision-making processes of regional mobility and their labour market outcomes. The starting point is the assumption that mobility decisions are not made in isolation, but are the result of a bargaining process, which is embedded in the respective regional as well as partnership context. Consequently, the decision to migrate is modelled from a dyadic perspective with multiple levels of explanation as well as multiple decision stages. Through the formal implementation of the decision-making process of spatial mobility in the form of a multivariate non-linear probit model with multilevel structure, it is possible to model the dyadic decision-making process differentiated in relation to the multi-level character (individual, household, and contextual level) of the exogenous migration determinants. As a result, the diverse characteristics of spatial mobility decisions, which have been identified in previous research, can be simultaneously estimated in a single model. Because of the distinction between actor-specific mobility dispositions and the mutual mobility decision, we are able to consider dyadic interactions as well as relative decision weights of the included actors. Therefore it is possible to incorporate the social process, which leads to the decision itself. Furthermore we are able to incorporate the complex mechanisms of local opportunity structures in the decision-making process of multiple actors through the estimation of contextual effects and individual-context interactions on the actor-specific mobility dispositions. In the second part of the research project, we inspect the labor-market outcomes of regional mobility. Here we are able to model the potentially varying mobility consequences of heterogeneous mobile agents because of the explicit distinction between actor-specific differences in labour market resources and the consideration of the actor-specific regional localization. As a result, it is possible to identify the key individual and context constellations, which determine the profit level of regional mobility and thus locate economically diverse successful groups of mobile agents. The Socio- Economic Panel (SOEP) provides the basis for the empirical implementations using small-scale structural features (SOEP geodata).