Child pages
  • Kosowsky Assignment 5

Julia Kosowsky

GIS Assignment 5

3/30/2014

Project Data Preparation and Basic Spatial Analysis

1. Folder Structure:

2./3./4.

For my project I am mapping the state of Kentucky. Therefore, an appropriate projection would be:

Projected Coordinate System:    NAD_1983_StatePlane_Kentucky_FIPS_1600_Feet

Projection:    Lambert_Conformal_Conic

Linear Unit:     Foot_US

I am also clipping all of my data to Kentucky

Below are my four data sets with their projection information, clipping, and areas (where applicable).

 

Kentucky Public Schools

Preprocessing:

     Already projected to NAD_1983_StatePlane_Kentucky_FIPS_1600_Feet

     Already clipped to Kentucky

Source: Kentucky Geography Network

Name: NAD_1983_StatePlane_Kentucky_FIPS_1600_Feet

Projection: Lambert Conformal Conic

Linear Unit: Foot US

Clipped Data:

 

Hospitals

Pre-processing:

     Projected differently (NAD_1983_Lambert_Conformal)

     Already clipped to Kentucky

Source: Kentucky Geography Network

Before PROJECT tool:

Name: NAD_1983_Lambert_Conformal

Projection: Lambert_Conformal_Conic

Linear Unit: Foot_US

After PROJECT tool:

Name: NAD_1983_StatePlane_Kentucky_FIPS_1600_Feet

Projection: Lambert_Confformal_Conic

Linear Unit: Foot_US

Clipped Data:

 

EPA Environmental Justice (Low-income, Minority, Low-income/Minority, Not Environmental Justice Community)

Pre-Processing:

     Already projected to NAD_1983_StatePlane_Kentucky_FIPS_1600_Feet

     Already clipped to Kentucky

Source: Kentucky Geography Network

Name: NAD_1983_StatePlane_Kentucky_FIPS_1600_Feet

Projection: Lambert_Confformal_Conic

Linear Unit: Foot_US

Clipped Data:

 

Area in Miles of EJ tracts:

 

Educational Attainment- Individuals 25 and older with High School degree or equivalent

Pre-processing:

     Projected differently (GCS_North_Amercian_1983)

     Already clipped to Kentucky

Source: ACS 2012 5 yr

Before PROJECT tool:

Name: GCS_North_Amercian_1983

Angular Unit: Degree

 

After PROJECT tool:

Name: NAD_1983_StatePlane_Kentucky_FIPS_1600_Feet

Projection: Lambert_Conformal_Conic

Linear Unit: Foot_US

Clipped Data:

** I struggled with this data set for a long time because the excel was a little off so it kept the joined attributes in string form so I could not use categories BUT eventually I got it to work only to see that one little census tract throwing everything off (over 300% of individuals 25 and older have a high school degree or equivalent...). I went back to the data and found that census tract in Meade county and this is what the data looks like. The red is the row for this census tract and the purple are the columns that I am using (total population over 25 and people over 25 with high school equivalency or high school degree).

**There are also a handful of census tract with no data. I think this is because there were four columns at the end of the excel spreadsheet that all had X’s instead of numbers and I was afraid that this was what was making it “string” so I removed them with advice from the TA. Did not know how to deal with this (so I held off in using this data with the tools below) but look forward to your advice as the project progresses!

Area in Square Miles of Census Tracts:

** It would not let me do this in the attribute table that had already been joined to the census data so I had to just do it on the plain unjoined census tract shape file (same distances but an interesting glitch?)

 

5. Using 4 Tools

**Because I do not know exactly what questions I am asking for my projects my questions are a little more general/exploratory

 

Summarizing Data in an Attribute Table

Although I am not sure what exactly my project is going to look at, one of the things I am curious about is what kind of access to healthcare students have. The state of Kentucky did not have a list of clinics, but it did have a data set of hospitals. I want to know what types of hospitals these are and how many of each type there are. Once I figure out my project I might also want to select some of these out (for example, PSY). One that might be particularly relevant are the CAHs which are “Critical Access Hospitals” common in extremely rural under resourced areas. Below is my summarized table:

Now I know what different types of hospitals, and how many of these hospitals are available in Kentucky. This could be out of date or hospitals could be mislabeled. Also, as mentioned above, students might have access to clinics or other doctor’s offices that are not included as “hospitals.”

 

Spatial Join

I am interested in environmental health and justice, and also in education so it is important to figure out which schools are in areas that are potential environmental justice communities. In order to do this I need to see how many schools are in areas that are either low-income or minority or low-income minority. To do this I first used spatial join to join schools (point data) to the EJ data set and got this new attribute table:

Now that I have this data set I can do a number of different things. For example...

 

Select by Attribute

I still want to know how many schools are in EJ regions. In order to figure this out I need to select by attribute for areas that are low-income, minority, or low-income/minority.

Success:

Or, in order to figure out how many schools are in the selected areas I can go back to Summarize and I get this helpful chart:

Now I know that there are 460 schools in Low-income areas, 133 in Minority areas, and 139 in Minority/Low-income areas. This could be wrong because of all the variables involved (schools could change or close, demographics of certain areas might change).

 

Near Tool

I also might be curious how close the schools are to the hospitals. For this I used the near tool because for each school, I want to know which hospital is closest.

 

Now I have the distances of schools to hospitals in Feet (because that’s the projection-- might want to change this considering these distances...) and I can match it to the FID from the hospital attribute table below:

This data could be incorrect because the near tool did not figure out which hospitals are closest to each school following roads or transport, so maybe some of these hospitals are not actually the “nearest” to access realistically. Also, I used all the hospitals, rather than selecting out any unnecessary hospitals (mostly because I have not figured out the focus of my project), which I will definitely need to figure out for the future. Again, these schools could also be close to some type of clinic that would offer necessary healthcare services, but which Kentucky does not appear to have the data for (I’m still looking!).