MOCO Map of the Week--Ethnic Clustering Part 2

Today's map is a slight variation on my last map, which showed the spatial distribution of the county's Black/African American population.  Today's map is a standard deviation map.  This means that each tract is categorized based on where the Black/African American share of the population fits in the wider distribution (i.e. within one standard deviation of the mean, within two standard deviations, etc.).


A bit on measurement:  In the tracts outlined in black the Black/African American share of the population that is greater than 2 standard deviations from the mean share.  The mean share of the Black/African American population is 16.62%.  The standard deviation is 13.58%.  Tracts outlined in black have a Black/African American share of the population that is greater than 43.78% (i.e. the mean + 2 standard deviations).  

So, what are the notable patterns here?  Unlike the earlier standard deviation maps I produced for the Asian and Hispanic populations (you can see them here), there are not multiple clusters.  Instead, all of the tracts with a statistically significant share of the Black/African American population are near one another and located on the eastern edge of the county.  In many ways the eastern orientation is not surprising given that Blacks/African Americans comprise a majority of the population in neighboring Prince George's County.  

Map of the Week Bonus Edition--Ethnic Clustering

This is for all you map lovers out there!

The last two maps I've made showed that the distribution of two minority populations--Asian and Hispanic--in Montgomery County.  Today, I thought I'd take a look at the degree to which these two minority groups are clustered in a statistically significant way.

Stay with me statistics phobes.  There are pretty maps involved!

One way to see evidence of clustering is to create a standard deviation map.  This means that each tract is categorized based on where the minority share of the population in that tract fits in the wider distribution (i.e. within one standard deviation of the mean, within two standard deviations, etc.).

Evidence of statistically significant clustering would be found in tracts where the Asian or Hispanic share of a tract's population is more than two standard deviations from the mean.  

For both groups--Asians and Hispanics--there a 12 tracts with significant spatial clustering.  On the maps below these tracts are shaded in dark blue and outlined in black (as opposed to gray).  For the Asian distribution this means tracts where Asians comprise 29.4% or more of the total population.  The comparable figure for the Hispanic population is 42% or more.

Although both groups have significant clustering, the clustering tends to happen on different sides of the county--north and west for Asians and south and east for Hispanics.

What to make of all this?  Montgomery county may be exceptionally diverse, but that diversity isn't evenly across space.


Map 1: Asian Share of the Population, Classified by Standard Deviation

Map 2: Hispanic Share of the Population