Photography by:
  • Machona James Ragedi, Clive Hassall, Susan Snaddon

Understanding different scales of social mixing

Populations are constantly changing in terms of their size and location. This means that assessments of segregation should be revisited, and done at various scales, in order to craft a complete understanding of the role that segregation plays in urban inequality. Understanding the scale at which segregation is the most severe would help to understand the type (or scale) of interventions required to improve residential integration and equality. Segregation indices tend to be low when calculated at large scales and analysing segregation only at a large scale would not reveal the effect that increases in belonging and exclusion at a local level have on the distribution of opportunity in the city. It is at local scales that it becomes evident how segregation limits access to opportunity.

This project is primarily aimed at a data driven output, supplemented with a brief literature review, which will represent social mixing in Gauteng at different geographic scales. The project will collate various inputs and analyses from previous and current GCRO publications into one output, introduce new data and analysis, and supplement these with selected in-depth case studies to illustrate the dimensions and implications of social mixing at different geographic scales. The literature and previous outputs that inform this project highlight the importance of understanding different dimensions of segregation in our society as well as the value of understanding spatial patterns of social mixing at different geographic scales. Each geographic scale that is considered has unique dynamics and barriers that shape residential segregation patterns and therefore provides valuable insights. The various analyses culminate into a ‘typology of diversification processes’ which is explained through various neighbourhood case studies.


Last updated: 22 August 2022.


The GCRO sends out regular news to update subscribers on our research and events.