>>> dhenwood at panix.com 05/10/01 03:00PM >>>
Charles Brown wrote:
>CB: I'm trying to think how they draw the census tracts and whether
>if drawn differently , there would be a different result.
They're meant to be pretty homogenous. <http://www.census.gov/geo/www/cen_tract.html>:
>Census tracts are small, relatively permanent statistical
((((((((
CB: Thanks for that.
I guess I am thinking that in analogy to political district gerrymandering , when a population's segregation/integration index with respect to subpopulations is based on drawing lines, it is possible to draw the lines such that things are very segregated or very integrated for the exact same residential pattern ( I am not saying though that the census tracts were drawn with the purpose of raising Detroit's segregation index , only that the pattern could be an accidental result of how the lines were drawn.).
I don't know if I can draw the picture, but if the lines are drawn chopping up "Black Bottom" ( in the 1940's with legally enforced segregation most Black people had to live in one contiguous area) , that old segregated residential pattern could have a high integration index. A similar thing could happen in reverse today.
I am glad to have this census analysis, but my comment is based in part on the fact that in Detroit there is no sense or common discussion or experience that "all the white people live in such and such an area" ( the way practically all Black people lived in Black Bottom). There are very few places or no places that I can think of that are referred to or thought of as "white neighborhoods". Even in the areas that have "more" white people than on average, there are Black and Hispanic people in significant numbers living very much mixed together. There are no all white or majority white schools.