How do we better measure Racially Segregated Networks and neighborhoods?

John Hipp and I and his student did a paper on the the spatial distribution network. They developed, and I kind of just helped, but they pioneered away of redrawing maps in a fashion that is informed by social networks. What we think of as a neighborhood, the census tells us what is a neighborhood, and we have mental maps of what neighborhoods are like, but this was another way of thinking about a neighborhood that was based on actual social ties. Where there was clustering in terms of both space and social space, they helped develop a technique of redrawing the map of neighborhoods based on that information. It looks different than census to find neighborhoods. It makes some more sense. Most of the time the Census does not have that information. They can't use that but it points towards some interesting potential where the social networks segregated in any particular way, there's quite a bit of variation in the racial segregation. Also there's a way of measuring segregation in a network by any category which ranges from one to negative one and it's all relative to the chance expectation. Zero would be the the ties are randomly assigned. They're no different than what you would expect them to look like by chance. And where one would be all of the ties are within category and negative one would be all the ties are cross category. So, if you're thinking about romantic ties in a purely heterosexual environment, that can be a negative one. Everybody's partnered up with somebody who is different from them. All the schools in the study all have something above zero, but there was quite a bit of variation in how socially integrated the schools are and and that is one of the questions I'm working on right now with one of my graduate students. We don't have the answers to the implications of that yet but one of the interesting things we do know is that however segregated racially segregated the friendship networks are...