An agent-based microsimulation framework for investigating residential segregation using census data

Social and urban structures are inherently complex and multidimensional. They do not arise from a single process, but from the multifaceted interplay of different social and economic processes. This is true for racial and residential/neighbourhood segregation which is the result of many factors, and its causes are not well understood. Over years, researchers have studied and measured the role and influence of different factors implicated in residential segregation, using traditional techniques and tools, mostly in social science. These include empirical studies and statistical analyses of survey and census data. Such complex and nonlinear systems are difficult to study analytically. To better understand the complexity of the social world, building a model – a smaller, less detailed and less complex version of the real world (structure, system) – is recognized as an indispensable tool. Simulation is an excellent way of modelling which results to a better understanding of social processes and features of the social world. Simulation can be used to test theories and the result of simulation will often be the development of further theory.