Introduction
When I was
still deciding on a major, I had considered choosing Environmental Public
Health because I was very interested in how people’s environments could affect
their health. Although I ultimately chose to pursue a degree in Biology, I
still believe that Environmental Public Health is an incredibly interesting
field, and one that can greatly benefit from the use of GIS. In class, one of our assignments was to watch
videos on the Geospatial revolution available through Penn State University. In the second episode, there was a segment on
food desert, and how maps could be used to make more compelling arguments to
address areas of greatest need. This seemed was very interesting to me and gave
me some of the inspiration for my own project.
Originally
I had thought of trying to map potential food deserts within Wisconsin but then
I had trouble finding free information on the location of grocery stores and
super markets. So I decided that I would
instead focus my question on exercise, or access to publicly available resources
that can help maintain an active lifestyle.
For the scope of the project, I simply wanted to map areas that would be
of potential interest for people working in public health. Areas, where limited access to certain
resources may cause people to live less active lifestyles.
When
generating my question for this project I had to consider the following:
What would my Area of Interest be?
Since I
currently live in Wisconsin, I wanted to choose counties within Wisconsin to
look at. While doing research on
Environmental Public Health I found articles that suggested correlations
between poverty and obesity. Links to a couple of these articles are listed below.
I decided that my area of interest would include 5 Wisconsin counties
with the lowest per capita income.
What criteria will be used in
determining a potential public health study area?
I wanted
to find an area that is lacking in resources that might encourage healthy
lifestyles. I decided proximity to
schools could be useful to consider because often schools will have tracks,
pools and other facilities that are open to the public. I chose to map hospitals for similar reasons,
and also because they often offer programs that promote physical activity and
healthy living. I also decided to
include parks and recreation areas because these seem like places where people
could go and be active. In the rest of
my writing I will refer to these areas as potential wellness resources, because
they are areas that have the potential to benefit people’s physical health and
activity levels. Obviously, not every
single place that is included in these areas will offer the potential resources
to the surrounding communities. It is
just likely that to some extent they do offer something that can benefit people’s
wellness. I also decided on an arbitrary
distance (5 miles), away from any of these resources that would be my proposed
study area, mainly because it is outside of a reasonable walking distance.
Question
In the
five Wisconsin counties with the lowest income per capita, what areas are five
or more miles away from any kind of potential wellness resource like a school, hospital,
park, or recreation area?
Methods
The first
step of my project was to find all of the data that I would need and to make my
data flow model.
The data
that I needed are listed below, along with where I got the data from.
Economic
information to determine Lowest per capita income counties in Wisconsin:
Downloaded comparative economic
characteristics 2013 American Community Survey 1- Year estimates
Location
of schools: ESRI 2013 U.S. geodatabase
Location
of hospitals: ESRI 2013 U.S. geodatabase
Location
of parks: ESRI 2013 U.S. geodatabase
Location
of other recreation areas: ESRI 2013 U.S. geodatabase
All of the
data that I used was from 2013, I believe that the data is timely and relevant. For the data downloaded from the U.S. Census Bureau,
I was relatively confident that the data was complete because it had
information for every Wisconsin county, however for the ESRI data I was not as
confident. When I used one of the ESRI
2013 databases to map cities I realized that some of the cities within the
counties I was looking at didn’t appear.
This presumably is because they were under a certain size. I was somewhat worried criteria that I was
looking at might not appear if they were too small. I did not find the concern reasonable enough
to not use the data because it was the highest quality that I could find. Also, when I added all of the layers to my
map, it did appear that the datasets were complete.
Once I had
my data I made a rough draft of a data flow model that I would use for the
project shown as Figure 1.
 |
| Figure 1. Rough draft of data flow model |
When I
actually began working on the project however, I realized that the model I had currently
set up would do a lot of unnecessary processing. After applying a 5 mile buffer to hospitals,
schools, parks and recreation areas I had planned to union them all together. This would have wasted time working with data
I was not going to look at in my final analysis. In my revised data flow model I decided that
after buffering the layers I would clip them using the 5 lowest income
Wisconsin counties, and then union them together. The revised data flow model can be seen in Figure
2.
 |
| Figure 2. Revised data flow model |
I followed
my data flow model closely during this project.
I started with U.S. counties and used select by attribute to select for
Wisconsin Counties. Then, I joined this
with the data that I had downloaded from the U.S. census bureau. After the
join, I went to the column that showed income per capita for each county and I
sorted it by ascending and then selected the first five listed and made a layer
from these. I then added US hospitals, parks,
recreation areas, and schools and put a 5 mile buffer around them. Once they were buffered I clipped each of
them using the 5 lowest income counties as a clip feature. I then used the union tool to make all of
these buffered and clipped features into one feature. After this was done I used the erase tool to
erase this area from the 5 WI counties.
What was left was a feature that contained the areas that are 5 or more
miles away from the potential wellness resources.
After this
was done, I worked on producing a map that was cartographically pleasing. This included adding a base map, removing
layers that made it too cluttered, producing a locator map, adding scale bars
and north arrows, and choosing colors that were easy to distinguish apart.
Results
The
results of my project can be seen in the map in Figure 3. I decided that I would show 6 different
frames with maps, and then have additional information about the map written to
the side of these. The individual frames
within the map were kept as simple as possible to compensate for the fact that
they are smaller in size. Because of the smaller size, the individual frames are also shown in figure 4-8. If it was
meant to be portrayed in a different medium, for instance a poster I would have
probably added additional details to the map. As it is now though, I chose to
show the locations of the counties within Wisconsin, and then the location that
I propose would make ideal public health study areas within these
counties. The proposed public health
study area being places that are more than 5 miles away from potential wellness
resources and located within the 5 lowest per capita income counties in
Wisconsin.
 |
| Figure 3. Maps with propose public health study areas |
Evaluation
My overall
impression of this project was that it was a great way to wrap up everything
that we learned during the semester. The
individual things that we were required to learn in previous labs were helpful
when working on our own spatial question.
If I had to do this project again, I think I would have started by
developing a better question. I would
want to do more research into environmental public health and see if there are
any characteristics of places that people have already expressed an interest in
researching. If I focused more on this
while developing a question, I think I could have made a project that had more
potential to be useful. I also think
that I might have considered using multiple buffers to show areas with differing
levels of concern.
Sources
US
schools, parks, hospitals, and recreation areas: ESRI
2013 US geodatabase
Comparative
economic characteristics 2013 American Community Survey 1- Year estimates:
U.S. Census Bureau: http://factfinder2.census.gov/faces/nav/jsf/pages/index.xhtml
Base map: Esri, HERE, DeLorme, USGS, Intermap,
increment P Corp., NRCAN, Esri Japan, METI, Esri China (Hong Kong), Esri
(Thailand), TomTom, MapmyIndia, © OpenStreetMap contributors, and the GIS User
Community
