Methods:
First statistical analysis performed was calculating the mean turnout and distance to nearest poll sites for each data frame. Districts with NYCHA housing have a turnout of 0.5 where ED without turnout at 0.55. Average distances were also compared, where ED with NYCHA housing are on average closer to poll sites than ED without by 0.05 km. KS tests were also employed seeing if the two types of districts behave similarly to each other with regards to turnout and distance from poll sites. KS stats are a good indicator of differentiating  turnout among voters, as seen in Borghesi et al.
An OLS linear fit was applied to the ED with NYCHA dataset, with distance to poll sites as the exogenous and turnout as the endogenous. A scatter plot does not tell much of the data and the so and OLS regression was applied to the data. Only about 2.9% of the variance can be described by the data. 
Conclusions:
It was determined that ED with NYCHA housing have lower turnout than those without, by about 0.05%. It was also found that the distributions between the two ED turnouts were different enough for significance.  ED with NYCHA housing are, by 0.05 km, closer to a poll site than those without NYCHA housing. The distribution of distance between the two ED were found not to be from the same parent distribution, which does say there is a statistical significance in the two's distances to poll sites (Figures 4 and 5). However, this experiment cannot say with confidence that distance is a result of respective turnouts.