how do you get those divisions with gray lines? i saw your NJ maps had one with municipality lines too
They're black.
The Census Bureau releases shapefiles for VTD's (or voting districts), which in most states correspond to election precincts. In Pennsylvania (or Philadelphia at least) VTD's correspond to ward divisions (I assume that these are not the divisions you are referring to). The Census Bureau also provides population data for VTD's, particularly racial data. The reason that the Census Bureau provides data on VTD's is to assist redistricting in states. In theory, if congressional, legislative, and other districts were assembled from election precincts, you would not have to to move voters into new precincts/polling places after the politicians finish gerrymandering. You simply distribute new ballots and tell the voters they've always been part of Eastasia, even though they were previously part of Eurasia.
In QGIS you can display a shapefile (the simplest way to do this is to drag the .shp file into the map pane. Each feature in a shapefile is represented by its geographical coordinates. For area shapefiles, this is a polygon (or if there are detached areas or islands, polygons). Each feature also has a set of attributes associated with it. These can be unlimited. For example, a State might have associated with it the capital, its population, legislative alignment, etc.
These attributes are stored in a .dbf file (Dbase file). A database may be characterized as a table, records, and fields (data base speak) or worksheets, rows, and columns (spreadsheet speak). QGIS knows that if you add somename.shp to a map, to look for somename.dbf associated with it.
You can join tables (add attributes to the feature). An easy way to do this is to save a spreadsheet as a .csv file (comma separated text), where one of the fields has the same value in both original .dbf file and the .csv file. The Census Bureau assigns unique ID numbers to each area, and these are typically used.
It happens that Philadelphia division names include the ward number (e.g. Division 10-12, is Division 12 of Ward 10). So you can create a spreadsheet with Division population data, and the ward number, save it as a .csv file and join it with the original .shp file (actually a copy of the original, since you probably want to preserve the file from the Census Bureau).
You can then display the division layer using the added attributes. For example this chloropleth was created displaying the ratio of Asian population to Total population in each division.
QGIS can also merge features based on some attribute. Since I had associated ward numbers with each division, I could merge the divisions into wards, creating a new layer. I can now display the ward layer in a number of ways,
Here random colors have been assigned by ward number. When I am producing a final map, I may have to edit color, so that there are not adjacent districts with different colors. In these maps, I have also thickened the boundaries. QGIS can also use different fill patterns, such as cross-hatches or dot patterns of different density.
This map was produced using Congressional district numbers that I had assigned to the wards. Other than the colors, the border thicknesses, the lack of labels, and the non-transparency, this is the same content as my third proposed division.
This map displays the ratio of black population to total population (black percentage) using 10 color levels.
In the maps that you had inquired about, I had set transparency to 50%, which permits the base layer (Bing road map) to display. This makes the colors less intense, and the borders grayer. Also, the contrast with the labels causes the borders to be perceived as being grayer. When I turn labels off, the borders are seen as being darker.
When I display a statewide map, I turn the borders off completely. Seen at a small scale, all ward boundaries make the area appear much blacker. In the maps above, Montco has county subdivisions, but I have disabled display of the borders (the portion of Montco attached to Bucks is outside this view).
Poll
