Schätzung komplexer Social Relations Model-Daten mittels Likelihood- und Bayes-Methoden

循环设计在许多不同的心理学学科中被用来研究人际感知、判断和行为的原因和后果。在这个设计中，每个参与者被要求根据一个人际变量来评判所有其他参与者，并且所有参与者也会根据这个变量来评判参与者。然后使用社会关系模型（Social Relations Model， SRM）对数据进行分析，得出许多不同的SRM参数估计。基于我们自己的工作，展示了如何使用似然和贝叶斯估计器来估计SRM参数，计划中的项目旨在将SRM扩展到更复杂的数据情况，并开发和实现用于分析这些复杂数据的似然和贝叶斯方法。例如，我们打算开发和实现多变量循环数据和纵向循环数据的似然和贝叶斯方法。总之，我们相信，计划中的扩展将使应用研究人员有机会使用适当的统计方法来调查一些新的和有趣的研究问题。

The round-robin design is used in a number of different psychological disciplines to investigate the causes and consequences of interpersonal perceptions, judgments and behavior. In this design, every participant is asked to judge all other participants concerning an interpersonal variable and the participant is also judged by all participants concerning this variable. The data is then analyzed with the Social Relations Model (SRM) resulting in a number of different SRM parameter estimates. Based on our own work showing how to estimate the SRM parameters with a likelihood and a Bayesian estimator, the planned project aims to extend the SRM to more complex data situations and to develop and implement likelihood and Bayesian methods for the analysis of this complex data. For instance, we intend to develop and implement likelihood and Bayesian methods for multivariate round-robin data and for longitudinal round-robin data. Altogether we believe that the planned extensions will give applied researchers the opportunity to investigate a number of new and interesting research questions using adequate statistical approaches.