There are (almost) no chain reactions in T.C. Schelling’s Checkerboard Model
Chain reactions occur in T.C. Schelling’s Checkerboard Model if and only if a discontent agent moves and thereby makes some of their previously content old or new neighbours discontent. While this seems analytically salient, it seldom actually happens in the model, meaning that usually only 10% of agents ever become discontent during the simulation. Moreover, chain reactions only marginally contribute to the _nal rate of segregation in the Checkerboard Model. Instead, the visually striking cluster structures at the end of a model run are already present in its initial situation as proto-clusters of initially content agents. Initially discontent agents attach themselves to proto-clusters of their kind and thereby cause high levels of segregation. Therefore, one should describe movement in the model as individuals` adaption to the initial situation rather than interaction between agents which would involve chain reactions.
Daniel is a junior lecturer at the Institute for Political Science, University of Bamberg since 2018; he pursues his Ph.D. in the field of Computational Social Modelling. He applies Computational Models mainly to questions in Political Epistemology, Collective Behaviour and Economics. Furthermore, he evaluates these models from an analytical and Philosophy of Science perspective.
Daniel holds an MSc in Social Research Methods form the University of Surrey, an MA in Ethics of Textual Cultures from The University of Erlangen-Nuremberg and an MA in Political Science from the University of Bamberg.
In his free time, Daniel enjoys building (allegedly) non-scientific models out of LEGO and extensive political discussions with friends, where he fancies argumentative rigor.
