Variations of the Segregation Model

The previous section explored the basic segregation model and the sensitivities of its parameters where it was demonstrated how neighbourhood sizes and preferences can result in different patterns of segregation emerging. This section will extend and refine the basic segregation model. Specifically the section will explore variants of the model which show how the effects of different assumptions or variations on the basic segregation model change the outcomes on the pattern of segregation.

Additionally, the following subsections will explore more generally some of issues when developing an agent-based model using geographical space and geometry. In a sense, this section could be entitled ‘experimenting with the segregation model.’ These experiments were carried out to highlight how the model behaves under different assumptions and conditions. Furthermore it allows for the testing of artifacts created in the model design, its implementation, and in the model running process. Each section will briefly outline why the model was created, and how it varies from the basic segregation model. The sections will then provide results, for the particular variations of segregation model and how the patterns of segregation emerge from the model.

These variations include:

• Addition and Removal of Agents
• The Density Model
• Movement to the Nearest Neighbour with No Prior Random Searching
• Agents Can Only Move in their Initial Area at First
• Preferences for Both
• A Polygon model of Segregation
  • Different Numbers and Different Types of Agents:
    • The Effect of Four Types of Agents
    • The Effect of Adding and Removing Agents
    • The Effect of a Small Minority Group
    • Agents have Preferences for Different Groups
• Combing Various Models Together

However, before exploring variations in the segregation model, a caveat is first needed. While numerous variations or extensions of the model are possible, the following subsections highlight how certain assumptions affect the outcome of the model and how changes can easily be made to the basic segregation model. Each of the following subsections in itself could be extended also to create further variations of segregation models. For this reason, each model identifies where the code for the model can be found, allowing future researchers to explore these models further if desired.