National Sample Survey of Registered Nurses

New nursing faculty on campus is researching various access to healthcare resources and possible correlations with the prevalence of particular nurse races/ethnicities. When I started discussing the data and spatial analysis my first thought was 'Holey-smokes...where the heck can we get our hands on data like this??'

Well, the best answer we have found so far is the National Sample Survey of Registered Nurses.

"Registered nurses sampled answer questions on their education and training in nursing, professional nursing certifications, education and workforce participation prior to becoming a registered nurse, current and recent workforce participation, income, demographic characteristics, and States in which they hold current licenses."

All sorts of great socio-economic data here. Very fantastic.

There are two public-use files available here from various years, 2000 being the most recent. We accessed the county-level 2000 file. Downloaded the zipped data to find an non-delimited ascii file, in addition to codebooks and a couple SAS scripts. We do not have SAS loaded in the GIS lab (to this point SPSS has been sufficient), so I made my way to another lab with SAS loaded and spent the better part of a morning building the SAS table with the provided scripts and then exported the data into a delimited format usable by ArcGIS. Then, of course, I ran it through my replacer program, one of the first programs I ever wrote. This little program uses regular expressions to remove all characters from the first line of a coma-delimited file except for alphanumeric characters (easily does this using the /w expression) so that ArcGIS can import the data. Then I imported the table into an Access database.

If anyone ever needs to do similar operations, here is the Access database I generated.

The website specifies the public-use county files identify the county and metropolitan areas that the nurses currently employed. However, as you can see in the database I generated, there is also a field for the city of current employment. I created a relationship in the Access database between the nursing sample data and the TIGER designated places table and got 7,246 matches! This means the 35,579 respondents in the public-use file can now be joined in ArcMap to particular 7,246 places!

<-- Here is a map showing the locations of these 7,246 places. As you can see, this forms a nice geographic distribution.

Now, concerning the race data that started this whole process, this datasets falls a tad bit short. I only found one field [RACE_GP] that contains the following coded values:
1 = "WHITE"
2 = "OTHER"
-9 = "UNKNOWN/REFUSED"
-6 = "DECEASED"

The professor is in the process of contacting the HRSA to see if they are willing to provide more data.

Either way, this dataset is a keeper and I am very happy to have come across it.

Google Maps: Mapping the Afghan Experience in the U.S.

Mapping the Afghan Experience in the US: http://gis.uta.edu/afghan
This site works best in any browser other than IE

Spatial Reserves

Our library's spatial reserves program just hit a new milestone as our latest project uses Google Maps as the front interface, as opposed to standard ArcIMS interfaces we have been using. (To learn more about our spatial reserves program, see this paper I presented at the 2004 ESRI Education Conference.)

Purpose of the Project

The purpose of Mapping the Afghan Experience in the US is to allow multiple sections (app. 20 sections) of a freshman English composition course (ENGL 1301) to explore various sociological aspects of the Afghan experience here in the US. To support our campusÂ’ 2006-2007 OneBook event, every freshman composition course is assigning students to read The Kite Runner, by Khaled Hosseini. A major portion of the book focuses on the character Amir (an Afghan) and his experiences in the US. Toward the end of summer, the information literacy librarian and I thought this provides an excellent opportunity to introduce GIS, mapping, and data into a freshman English composition course. Not only were we going to build a spatial reserves project for these classes, but we were going to use the Google Maps interface as the front end.

We presented the idea of to a number of the instructors, and they loved the idea of requiring their students to use both the traditional literature resources to write their Kite Runner papers as well as demographics gleaned from an easy-to-use mapping interface. I now have 20 instruction sessions with freshman English courses. For me this is a really, really big deal. I so very much want this to be successful as this can be the beginnings of a formal adoption of the use of data and GIS mapping in a required campus-wide course. If this program can grow, perhaps there will soon be a day when every single student is exposed to GIS as a freshman. Very cool indeed!

Why Google Maps?

Google Maps was used for three reasons:

1. Students are already very familiar with the Google Map interface. The idea is that their comfort level will be much higher with this interface than the standard ArcIMS interfaces.
2. The appeal and draw-factor of implementing a Google Map mashup will prompt students to explore the application and perhaps increase interest in GIS.
3. The availability of the base map and satellite/aerial imagery.

Mapping the Afghan Experience in the US

According to the Census 2000, there were only 22 counties with at least 100 Afghan residents. Users are presented with a Google Map zoomed to the 48 contiguous states (excluding Hawaii & Alaska) and a buffers marking those 22 counties. Users can then change the background map to different layers showing various demographic attributes, derived from Census 2000, establishment count fields, and layers downloaded from NationalAtlas.gov. Geocoding is also available. Metadata records are viewable for all of these layers. Users can also add markers for a few establishment types.

So, WhatÂ’s Under the Hood?

This ASP.NET application was developed in Visual Studio 2005. The application uses the Google Maps interface to bring together the following services:

• Google Maps API
• Public ArcIMS Service: Server: magellan.uta.edu, Service: afghanUS
• Point features in XML format

Thank You

A big THANK YOU must go out to the following folks who helped us to create this application:

• Jeremy Bartley of ESRI (formerly of the Kansas Geological Survey and Mapdex) was invaluable to us. He was very generous with his time, advice, and ColdFusion source code. Without Jeremy, we would not have been able to implement the Google Maps + Cached ArcIMS logic.
• City of Rockford, IL who created and released the following ASP source code for overlaying ArcIMS services with Google Maps. Here is the post concerning this on the MapDex Blog. This was invaluable to us as it made it much, much easier to adopt JeremyÂ’s logic into ASP.NET after viewing the classic ASP version.

Pizza for Intro to GIS Tonight

I was asked to teach an Intro to GIS class tonight for a professor who is unable to make the class. I think this is great. Much better than ripping these students off, and cutting them out of a class when the class only meets once a week. It's also not too bad for me as I taught this course last spring and summer.

The class is going over chapters 4 and 5 in the Getting to Know ArcGIS book. It's tough to just jump in for one class, but I want today's lesson to work well nonetheless.

So here are my ideas: Chapter 4 is 'Exploring ArcCatalog' and chapter 5 is 'Symbolizing Features and Rasters'. Not very exciting stuff on the face of it, but it's my job to amend that, right? I think we will go through the geoprocessing version 'Where to Open a New Pizzeria in the D/FW Metroplex' workshop that I devised a couple of years ago. This is a nice exercise that contains all of the new material in these two chapters.

Here is an overview of the steps we will complete:

1. Download and extract the personal geodatabase (data must be extracted to C:\ drive)
   
2. Explore the feature classes in ArcCatalog
3. Add the feature classes to ArcMap, playing around with symbology
4. Run first geoprocessing model (created using the Model Builder)
1. This model first calculate euclidean distance and convert features to raster.
5. Explore raster grids, playing around with symbology
6. Run second geoprocessing model
1. This model presents the users with a simple interface to enter weights and reclassification schemes for each variable considered. The model then ranks the suitability of each raster cell.

We should complete this exercise within 2 hours, after which I will end class.

Night of Linux Ends with GIS

Spent last night installing Linux OS on my laptop. Actually did it, and finished the evening off by creating a choropleth map (graduated color) displaying the distribution of median household income by block group in Dallas County. I used qGIS for this. (See left for screenshot.)

This also is my first blog post written in a non-Windows environment. Tastes a bit different, eh?

Here are the highlights of what I had to go through did to make this all happen.

1. Install Xandros Open Circulation, a Debian style of Linux.
1. This was extremely easy, which was why I opted to use the Xandros package. I repartitioned my hard drive to make a partition for Linux.
   
2. Set up my wireless connection.
1. Believe it or not, this was the most difficult step. My wireless network card, a Broadcom 802.11b/g wlan, must not be compatable with Linux. I had to use ndiswrapper to use the Windows drivers. Fine, I then had to find those drivers. Throw in a couple of reboots and retires, and it was a few hours spent setting this up. I eventually got it to work, and am writing this post on my wireless connection.
3. Install/update software using Xandros Networks, a Debian package management application.
1. I was very happy to install the latest version of Firefox at this time. Along with my wireless network, it is vitally important that I contine to have acess to particular Firefox extensions, such as the Yahoo! Toolbar to access my remote bookmarks. Web applications make this transition from Windows so much easier as there is less to install. Yahoo for email, Meebo for chat, Bloglines for rss, Easynews (paid) for usenet, the list can go on and on.
4. GIS time.
1. I know that qGIS operates on multiple operating systems, so I fire up Firefox and on my wireless connection I head over to qgis.org. The Linux version, I had to remind myelf as I headed over to the download page. They offer a Debian package. I download it and attempt to use Xandros Networks to install it. Wasn't meant to be as it could not locate the necessary dependencies. But the qGIS site also mentions apt-get as an easier installation method. Research apt-get, and it seems to be similar to Xandros Networks. Then I see that Xandros Networks has a search. I search for qGIS, and sure enough it finds the package. It installs easily and beautifully, just as the qGIS site advised it would.
2. So, I thought I would attempt to use qGIS to complete a basic exercise that my real estate class completed during the first class (see image above). I do not know how to do a tabular join using qGIS. Do I install PostGIS or MySQL? No, not today. I simply edit the shapefile's DBF file in Open Office Calc, and paste in the attributes (after sorting them, of course). Very impressive how well Calc was able to edit the DBF file. Excel does not work so well with DBF files. Very nice. Easy to create themap from there.
   
3. Well, the keyboard print screen button did not work. A screen capture program that was included in the installation, and this created the image you see above.

This was my first night's foray into Linux. As I accomplish more, I will post about it here.

Library GIS Events Fall 2006

Here is a list of the workshops, ESRI webcast seminars, and event booths that our library's GIS Program will participate in this Fall 2006 semester. As the dates get closer, I will post more details about particular events here.

As you can tell, workshops are my primary outreach method. I focus the workshops on how GIS and the analysis process can solve a particular problem. This is as opposed to focusing on the GIS software in general.

Workshops (hands-on, no experience necessary):

• Friday Night Hangout
• Thursday, September 21, 3pm - 5pm
• Central Library B20 (basement)
• Use suitability analysis to identify the best places to hang out this Friday night.
• Add directly to Outlook
• Create Your Own Google Map Mashup
• Thursday, October 12, 3pm - 5pm
• Central Library B20 (basement)
• Learn how to use GIS to easily create your own Google Map mashup.
• Add directly to Outlook
• Low Income Housing and GIS
  
• Thursday, October 26, 3pm - 5pm
• Central Library B20 (basement)
• Use GIS to explore the relationships between the availability of low income housing and various socio-economic indicators.
• Add directly to Outlook
• Mapping the Afghan Population in the U.S.
• Thursday, November 9, 2pm - 4pm
• Central Library B20 (basement)
• Find out where in the U.S. there are large centers of Afghan population. In the D/FW Metroplex, perhaps? This workshop is designed to support the UT Arlington One Book Program. This year's book is Kite Runner, by Khaled Hosseini.
• Add directly to Outlook

ESRI Seminar Webcasts:

• Getting Started with Scripting in ArcGIS 9
• Thursday, November 16, 2pm - 330pm
• Central Library B20 (basement)
• This will be a public showing (staff and students only) of this fantastic webcast from ESRI. Basic familiarity with ArcGIS is recommended.
• This is not a live viewing.
  
• Add directly to Outlook

Event Booths:

• Welcome Week Information Fair
• Tuesday, August 29, 11am - 1pm Central Library B20 (basement)
• Come over the GIS Lab, Central Library 510, to play around with the following GIS applications:
• PlateTracker: Bring any shapefile and this ArcMap extension will reposition the features in the millions of years backward or forward in time, based on tectonic movements. This extension was developed by Dr. Chris Scotese and Joshua Been.
• Friday Night Hangout: Use this Google Map mashup to find out where you and others should hang out this Friday night. So much fun. Learn about other Library Welcome Week activities here.
• Add directly to Outlook
• UT Arlington Technology Fair 2006
• Wednesday, November 8, 9am - 3pm
• E. H. Hereford University Center
• Come celebrate GIS Day one week early this year and visit the GIS booths at the Tech Fair! At the Tech fair, we will unveil our latest web application, Texas Time Machine: Mapping Our History. This is a mashup of scanned and digitized historic Texas maps from the Library's special collections. The raster maps and feature classes are served by ArcIMS, and the ArcIMS service is integrated into the Google Maps interface.
• Add directly to Outlook
  

Low-Income Housing Data & Social Work

Have not posted in over a month. Miss blogging immensely. Been a bit busy teaching 3 credits during the summer, teaching 6 credits this fall, consulting on some Python scripting on the side, submitted a paper for publication, and of course my full-time job as the GIS Librarian. Nuff said...

Low-Income Housing Data

Social work faculty and grad students have been over to see me frequently during the summer and now into the fall semester. They are primarily seeking assistance with collecting datasets and with a very high curiosity about how GIS can help them with their research. These folks are thrilled to work with the most readily available resources, such as Census data, AGS estimates, and facility locations. However, another frequent request is for low-income housing data. I had to do some digging, but here are the best resources that I have been steering them towards and implementing within GIS and SPSS. Before I prepare a formal document outlining the numerous low-income housing datasets, I want to flesh out my ideas here.

• HUD's LIHTC (Low-Income Housing Tax Credit) Database
• This database provides a complete list of all rental properties that receive funding through HUD's LIHTC program through 2003. These rental properties have program rent and income restrictions.
• Project and financial variables are provided for each rental property.
  
• Here is a link to the entire database in DBF format. Most records include lat/long, so this takes the pressure off having to geocode.
• Here is more information on the LIHTC program.
  


• Qualified Census Tract Table Generator
• As I understand it, a rental property must be located within a QCT (Qualified Census Tract) to qualify for LIHTC.
• This table generator creates tables specifying the FIPS code for tracts designated as QCTs in 2005 and 2006.
• This resource also generates tables specifying the tracts designated as a DDA (Difficult Development Area), which seems to be another way for a property to qualify for LIHTC.
  


• Residential Finance Survey
• Taken each decennial census since 1950, the RFS (Residential Finance Survey) is the only survey designed to collect and produce data about the financing of nonfarm, privately-owned residential properties.
• Provides financial data for multi-unit rental properties, including property and mortgage characteristics. Also includes renter information. Great stuff.
  
• The 2001 RFS Public Use File for both renters and owners can be downloaded in SAS format here.
  


• Special Tabulations Retrieval System
• Creates tables that provide counts of households by tenure, by income intervals, by age of householder, by size of household, by housing conditions as of the 1990 Census and the 2000 Census. These variables are available by themselves via Factfinder, but not cross-tabulated as this system provides.
• Downside is that this data is only available down to the county, MSA, and place level.


• HUD Boundary Files Download Site
• For a number of datasets available from HUD, census tracts and block groups are split by jurisdiction boundaries. This means that those TIGER and TIGER-derived files do not cut it here. Never fear, though, because this is where you can download these split tracts and block groups in shapefile format.


• Low and Moderate Income Summary Data
• As I understand it, families must be classified as earning either low or moderate income to qualify for any housing subsidy programs. This resource specifies the number of families that are categorized as low and moderate income down to the census block group level.


• CHAS 2000 Data
• This datasets is a special tabulation of Census 2000 designed to be used by local governments for housing planning as part of the Consolidated Planning process.
• Here is a list of the many, many variables available in this dataset. What a hidden, little known jewel this is!
• Data is available down to the census tract level for all areas, and the block group level for some areas.
  


• Picture of Subsidized Households
  
• OK, perhaps I saved one of the best resources for last. Or perhaps this was just the last item added to my bookmark list...
• This dataset provides, down to the census tract level, housing totals, public housing, section 8 data, section 236 data, other HUD subsidy programs, and LIHTC data.
• This is the only dataset I have yet found that provides section 8 data on a geographic level as local as a census tract.
• Down side to this dataset is that it is only available for 1996 through 1998. There is also a file that describes this data for the 1970's as well. But of course, if you contact HUD you can get more current data...
  

These datasets are forming the basis for a workshop I am holding this semester. I will discuss the four workshops I have planned in my next post, which will be sooner than 6 weeks away...