Olympic Changes in
Atlanta
Discussion
This discussion section looks at:
Social Index
A visual
examination of the change in the Social
Index map (results section) does appear to indicate that there are
areas close to the venues downtown where the Social Index has
increased, especially directly east of the venues. However, there
appear to be areas close by where the Social Index has remained rather
steady or decreases. We ran a correlation (see graphs in results
section), to see if distance from the venues has an effect on the
change in the Social Index, but there does not appear to be a
correlation at the Atlanta city (proper) level. We also calculated the
area of polygons at the census block groups for the each level in the
change in the Social Index at five increasingly large buffers around
the venues, and divided by the area of the each buffer (see Social
Index buffer oprecent change in the results section). There was a
slight drop off in the one to ten percent increase in the change in the
Social Index between 1000 and 2000 meters, there was liitle to no
change after. These results seem to indicate that although there have
been both increases and descrease in the Social Index around Atlanta,
the effect of venues on surrounding neighborhoods are inconclusive. We
also conducted a kernel density and hotspot clustering anlaysis (see
results section for maps) on the change in the Social Index. The kernel
density map does indicate that there is an area close to the venues
with a substantial increase in the social, however, because this map is
created from aggregated census data, one most be wary of reading too
much into the kernel density aanalysis map. Also, the hot spot
clustering anaylsis does show clusters of Social Index increases near
the venues.
Property Values
Again, visually there does appear to be evidence
that some increases and perhaps decreases in the median property value
are associated with the location of venues (see maps in the results
section). We used the same techiques as the above Social Index
analysis (except we did not do a kernel density or hot spot cluster map
for increase in property values). The analysis appears to indicate the
proximity to venues has little effect, overall, on property values,
even though there are areas of both increases and decreases near the
venues. Essentially, the median property value does not decrease or
increase further away from the venues.
Land Use
The maps and table from results indicate that the
Olympic venues did change the land use significantly in the areas
around it. Visually, the buffered areas around the Olympic venues
show a significant change between 1980 and 2003. The graphs
comparing expected land use changes and observed land use changes also
indicate the significant increase in commercial and services land
use. There has been an increase since the 1980's and this is
especially significant in the 1 km buffered area around the Olympic
venues. This may indicate that the Olympic venues had a localized
effect on land use.
The
chi squared values calculated for the 3 different buffer distance show
that the venues affect the land use significantly. The chi
squared values for the 1 km, 3 km and 5 km buffered areas are 378601,
6816977 and 2513204 respectively. There is 3 degrees of freedom
for this calculation. Thus, the Olympic venues were hugely
significant in land use changes.
One thing to keep in mind is that the mismatched
coordinates for different land use years may have skewed the data
set. The Olympic venue buffers were mapped using the 2003 land
use data set and these coordinates do not match the 1980's data
set. This affects the 1 km buffered areas the most because the
buffers are not located exactly where the venues should be based on the
1980's map.
