Lesson 6

Design/Build GIS (Week 3)  -  Analysis and Display

 

Deliverable Set 1:

 

Figure 1:  Voter Registration Percentages by Census Block

 

Figure 2:  Hispanic Voter Registrations by Census Block Showing Registration Count

 

Figure 3:  Party Affiliations by Census Block

 

Figure 4:  Closer View of Hispanic Voter Registrations and Count Compared to All Voter Party Affiliation by Census Block

 

 

Figure 5:  Registered Voter Percentages Compared to Total Party Affiliation by Census Block

 

 

Figure 6:  Percent Voters Table

 
Deliverable Set 2:  Questions to be Addressed.

 
1.  What areas of the city have the lowest percentages of registered voters?

From Figure 1, we can see that the eastern-most and the western-most census blocks have the lowest percentages of registered voters.

 
2.  Is there any variation in party support within the city (i.e., do different parties dominate different parts of the city)?

A close examination of Figures 3, 4, and 5 above, indicates that the area around the central to south central portions of the city tend to favor the Whigs while the more outlying regions of the city tend to favor the Mugwumps.

However, from a voting perspective, I wanted to know why this might be the case.  In an effort to find this out, I exported the data table that contained the registered voting data and the corresponding census data.  I found that I could use Microsoft Excel to manipulate the data in the table, adding appropriate fields and performing calculations, and then once I saved just the newly calculated data could save this as a new .dbf file and then join this data table to an existing data table within ArcView to derive a couple of new data frames.  Figures 7 and 8 below represent my newly calculated information, Persons per Square Kilometer and Houses per Square Kilometer.  I wanted to calculate these values to set up an examination of voter registrations based upon housing and population density.  My hypothesis was that the Whigs tended to have higher voter registrations in more densely populated areas and the Mugwumps tended to have higher voter registrations in less densely populated regions.

Figure 7: Population per Square Kilometer

 

Figure 8:  Housing Density per Square Kilometer

From these two layers, an obvious population density picture emerges.  In the south central region of the city boundaries, two major population centers are found with tentacles reaching out to the south and to the east along what may be major transportation corridors.  When these images are overlaid with the voter registration information as is shown in Figure 9, it becomes more obvious that, based on voter registration data, the urban regions tend to favor the Whigs and the rural regions tend to favor the Mugwumps.

Figure 9: Closer View of Party Affiliation by Housing Density

This analysis is not perfect, as this observation is not an iron clad rule, but for the most part the central core area of the city as represented by greater housing density does seem to have a slight advantage in Whig voter registrations.  Also, because of the greater housing density, the number of registered voters tends to be slightly higher then for the surrounding, less densely populated areas.  Interestingly, the Whig advantage is still not sufficient to overcome an overall city-wide advantage to the Mugwumps.

 
3.  Should the Whigs invest in Spanish language flyers for the growing Hispanic population?

First, and not to be too flip about it, I'm not sure there is enough information provided to answer this question.  For example, how do we know that the Hispanic population is growing.  To test this, it would seem that we might have to look at several decades worth of census data to see if (a) the Hispanic population is indeed growing, (b) if so at what rate, and perhaps most importantly from a voting perspective (c) where is the largest growth occurring.  However, let's for the moment assume that the Hispanic population is indeed growing, it's not a question of whether the Whigs should invest in Spanish language flyers, but how many should be made and where should they be distributed?

Figure 10 below shows the density of the Hispanic populations overlaid with the city's voter registration patterns.  Where there are already a high number of Hispanics, the Whigs tend to have more party registrations.  Further analyses would need to be conducted to determine if Hispanics are predominantly registered as Whigs - but for the purposes of this analysis let's assume that there is that high correlation.

Figure 10:  Hispanic Population Density vs. Overall Party Registrations

Armed with the knowledge that the Hispanic community is largely supporting the Whigs, I would definitely invest in Spanish language flyers.  I would target the areas where there already is a high Hispanic voter registration to reinforce the Whig party commitment to Hispanic issues, but I would really push to distribute large numbers of flyers in those areas where there is a moderate number of Hispanic voters but where the Mugwumps enjoy a voter registration advantage.  If, as we assumed to be the case, the Hispanic population is growing and that Hispanic voters tend to favor the Whigs, these moderate to high Hispanic areas with higher Mugwump registrations can become the areas targeted by the Whigs to make significant inroads into the overall Mugwump voter registrations advantage.

Alternatively, if the purpose of the flyers is to promote voter turnout, then it would seem that you would want to distribute those flyers in the areas where they will do the most good, i.e. the areas that have the highest Hispanic voter registration percentages.  That is exactly what Figure 11 displays.  The darker areas on the map are the areas that have the higher percentages of Hispanic voter registrations.  This was calculated by adding a field and using it to divide the total voter registration values for each census block into the Hispanic voter registration count.  Multiplying that by 100 provides a percentage value that is shown on the map in 5% increments.  Using this analysis, you would want to distribute the flyers in the census blocks denoted by the darker colors.

Figure 11: Hispanic Voter Registration Percentages

 
A final look at this issue of where to distribute flyers for Hispanic voters is to compare the Hispanic population density with the percentage of registered Hispanic voters.  This analysis was based on the same percentage calculations made for Figure 11, but instead of displaying a map, just the label values for the registration percentages are shown (transparency = 100% for the data frame).  This was then overlaid onto the Hispanic population density map.  In theory, the higher the Hispanic population density should result in higher Hispanic voter registrations.  Where there are low Hispanic voter registration percentages in a higher Hispanic population density, these census blocks then become targeted areas to drive potential Hispanic voters to register.  Such an area is found throughout the central areas on the city in Figure 12 below.

On the other hand, if the flyers are to get already registered Hispanics to the polls, then you would need to target those areas where there is a high percentages of registered Hispanic voters.

Figure 12:  Hispanic Population Density with Hispanic Voter Registration Percentages