Ethnic residential segregation is an enduring feature of American cities. Residential location and neighborhood context are important factors that affect the life chances of individuals, families, and minorities. Two problems hinder general understanding of segregation patterns and the forces that shape them. First, the research literature relies on technical, quantitative indices computed from specialized data sets to document segregation patterns and trends. Second, conventional methodologies for exploring effects of hypothesized determinants of segregation (market forces, the spatial and demographic organization of cities, household preferences, constraints imposed by discrimination, etc.) require advanced data analysis skills. To overcome these problems Mark Fossett of Texas A&M University authored SimSeg, a computer program that uses intuitive graphical presentations to help students comprehend segregation patterns and simulation capabilities to help researchers explore different theories about the determinants of segregation patterns. In Phase II, Amber Waves Software and Texas A&M will refine and extend the SimSeg model by: 1) improving its ability to represent the spatial and demographic structure of American cities, 2) enhancing its model of household residential choices, 3) enabling the model to represent any set of ethnic groups, 4) developing tools that will enable students, researchers, and social science professionals to build their own simulation scenarios, 5) working with outside sociologists to ensure the scientific merits of the SimSeg model, and 6) conducting extensive internal verification and external validation exercises. Additionally, Amber Waves Software and Texas A&M will build a library of simulation scenarios based on real cities and develop extensive technical documentation of the SimSeg model.
| Fossett, Mark (2011) Generative Models of Segregation: Investigating Model-Generated Patterns of Residential Segregation by Ethnicity and Socioeconomic Status. J Math Sociol 35:114-145 |
