mbrown7
commented
9 years ago Aspects of segregation:
Evenness: underrepresentation and overrepresentation in areas Exposure: limited interaction with majority members Concentration: occupying less physical space than majority members Centralization: congregating around "urban core" and more central than majority members Clustering: tightly clustered versus scattered
"overlap empirically... conceptually distinct"
I don't know that the physical measures like being centralized would apply here, since we're not using a CA or something else utilizing physical location. We could expand the idea to relate to things like student housing, though - so it could be a principle we consider, but a direct implementation obviously wouldn't make sense here.
Measure of evenness: dissimilarity index. "It measures departure from evenness by taking the weighted mean absolute deviation of every unit's minority proportion from the city's minority proportion, and expressing this quantity as a proportion of its theoretical maximum (James & Taeuber 1985)." This is basically what we did for our measure of similarity, which is neat. "It represents the proportion of minority members that would have to change their area of residence to achieve an even distribution, with the number of minority members moving being expressed as a proportion of the number that would have to move under conditions of maximum segregation." It also offers some equations and derivations/support for them.
Measure of exposure: "degree of potential contact." Measures "experience of segregation" instead of measuring some departure from ideal "evenness." Two indices: a) "The extent to which members of minority group X are exposed to members of majority group Y, and it is usually called the interaction index. It is the minority-weighted average of each spatial unit's majority proportion." b) "The converse of the interaction index is the isolation index, which measures the extent to which minority members are exposed only to one other, rather than to majority members, and it is computed as the minority-weighted average of each unit's minority proportion:" This seems like evidence for using "proportion of encounters which are with minorities" instead of "proportion of successful encounters which are with minorities" since we are interested in how "able" you are to encounter people of the other race, not just make friends with them. This is really useful information, in my opinion, and makes me want to change the metric for how I was measuring change in segregation.
Measure of concentration: This has to do with spacial locations so I'll skip the metric they propose.
Measure of centralization: This is also based on spacial location so I'll skip this, too.
Measure of clustering: This is also based on physical loction.
In the end I think this is a really awesome paper to cite, because it gives support for a lot of the aspects of our simulation that we put in intuitively. For instance, the fact that students meet each other based on their groups (a little bit randomly too, but focus on groups) ties into the idea behind exposure, that who you are exposed to contributes to your segregation/lack thereof. This gives support to our idea that a main way to experience and/or overcome segregation depends on who you are being exposed to (who you are encountering).
divilian
Skim Massey and Denton 1998, figuring out their "5 ways to measure segregation." Is one of these the best for us to use to quantify our d.v.?