Rich Internet Application (RIA) Technology Will Help Reveal Patterns in 2010 Census Data
By: Ross H. Capaccio
President/CEO, röös design + consulting
Technical Level of Article: Beginner
Definition: Rich Internet applications (RIAs) are web applications that have most of the characteristics of desktop applications. Rich Internet Applications typically overcome the limitations of browser incompatibility by utilizing cross-platform development environments. For right now, don’t worry about this. In part two of this series I will go more in-depth on the technical details; right now I just want you to understand that this technology exists. (From Wikipedia, the free encyclopedia keyword search = RIA)
Definition: A Geographic Information System (GIS) is computer software that links geographic information (where things are) with descriptive information (what things are like) presenting many layers of information on a map. (1) GIS is perhaps the only information technology that requires a major digital infrastructure referred to as basemaps. (2) Most, if not all of these basemaps are built and maintained by government and commercial companies. (1) From ESRI, “Mapping for Congress: Supporting Public Policy with GIS” (2) “GIS Tutorial for Health, Third Edition.”
_______________________________________________________________________________________________________________
New insights
Large corporations, government agencies and membership based organizations will soon start preparing to use new information collected during the upcoming 2010 United States census. The U.S. Census Bureau’s 2010 census is expected to reveal changing demographic patterns resulting from large displacements of people due to natural disasters and disruptive economic trends that have occurred over the past decade. For many organizations who are entering into a new strategic planning cycle this is good news as this new data will help them define their existing operating environment and for establishing a baseline for measuring change over the next half-decade or so (a common planning horizon). Every ten (10) years with the arrival of this new data come new opportunities to transform the data into compelling and actionable knowledge using the currently recognized state-of-the-art technology. This decade, Rich Internet Application (RIA) technology will be one key technology for making the information easily available.
In this blog I will show how operational data can be combined with census data to create new insights into the data by “visualizing” the data in new and exciting ways using RIA technology. The example illustrated in this blog will take addresses of public libraries from the National Center for Education Statistics (NCES) Public Libraries Survey, Fiscal Year 2005 which can be found at http://nces.ed.gov/surveys/libraries/ and place them on a RIA Web map. This demo only uses library locations in Cook county Illinois, counties surrounding Boston and counties surrounding San Francisco. In addition to mapping the address, this demo explains in general terms the concepts (this is not a tutorial on using GIS software) on how to geographically append other business critical data to a database or spreadsheet file. In this demo I added the corresponding census tract code and select census variables to the original NCES Excel file using GIS. After the census code is appended, any census variable can be added and analyzed and then visualized on a map, say in PDF format, or through a map-centric dashboard application using RIA technology. In this example I use an Adobe Flex application based on the ArcGIS Flex API from Environmental Systems Research Institute (ESRI), Redlands, CA . In part two of this blog I’ll talk about more techie stuff like APIs, etc.
Sharing the knowledge – Preparing the Data
Each new census is also a time where the latest information technologies are showcased and looked upon to be used to analyze report and distribute all this new data. For the 1990 census it was CD-ROM and for the 2000 census it was Web portals that allowed you to slice and dice census data using American Fact Finder and them download the data and receive an Excel or text file as an end product or view a static map. If you wanted to map the information you usually had to find a GIS expert to create a custom view of the data to meet your needs, or you used paper maps provided by the Census Bureau. In either case it was a difficult time consuming process.
For the 2010 census the enabling technology will be Web services for hosting the data and RIAs for viewing the data. These technologies, are good at explaining the “what,” “who” and “when” about the data, but not the “where.” Columns of numbers on a data table don’t tell the whole story. For a more complete picture, an organization needs to combine operational data with geographical information to display visual patterns and trends across areas. This is the role of geographic (geospatial) data.
An often cited fact is that eighty (80) percent of all databases and spreadsheets contain a “where” or location component in the data, usually a U.S. postal address, a zip code, or the name of a county. It’s the location component that opens up new possibilities for data analysis and visualization. The maps in this exhibit were all generated using a standard U.S. postal service address: “Street number and name,” “City,” “State,” and “Zip code” found in an Excel file. Here is how it’s done.
Preparing the Geographic Data
A GIS process known as “Address Geocoding” was used to convert each address in the NCES Excel file into a corresponding latitude-longitude pair. The program ArcGIS was used to do this. The NCES Excel file now contain two new fields, X and Y coordinates. This allows the placement of symbolized points on a map. In this example the blue library symbol was used as seen in Figure 1.
In this exhibit the rows and columns of our spreadsheet come to life just by knowing how to leverage the “where” component in a typical spreadsheet.
Figure 1
Give the application a few seconds to load. You’ll see three clusters, Boston, Chicago and San Francisco. Use the following instruction to navigate the map.
Here are a few suggestions for navigating the RIA map.
Map navigation using mouse:
- Press Shift and click map to recenter on clicked location
- Double-click map to recenter and zoom in at clicked location. To zoom out, press Alt when double-clicking
- Use mouse scroll forward to zoom in at current mouse location
- Use mouse scroll backward to zoom out at current mouse location
- Press Shift and click and drag to draw a rectangle around an area and zoom in. To zoom out, press Alt and Shift when drawing the rectangle
After the points were placed on the map, additional information contained within the ECES spreadsheet was used to create three (3) custom views of the data in three map themes seen in Figure 2 below: (1) Total Expenses, (2) Funding Sources, and (3) Library Staff Composition. Each theme is symbolized using a pie chart but many other methods could have been used, including 3-dimensional data mapping techniques. Download a PDF version by clicking the image (7MB).
Figure 2
A second, very powerful GIS process known as “Geoprocessing” was used to manipulate the Excel data even further, adding new information to the existing list of attributes collected by the NCES. In this example I performed a spatial join of the NCES data with census tract data. Spatial join is one of several hundred geoprocessing operations that GIS can perform. To illustrate how geoprocessing can append new information to this file various census variables were added at the census tract level. Go back to the library location map http://www.roosdesignconsulting.com/bin-release/simplepoints.html# and double click on a point and you’ll see that census data was appended to the original NCES file. I also added a new field to the NCES Excel file containing the name of the closest hospital to a library. Today, public safety and homeland security concerns are of interest to community 911 systems, so if someone requested that this information be added to the database, or any other data for that matter, the spatial join geoprocessing function could be used.
Sharing the knowledge – Building a Web Map Centric Data Viewer Dashboard
We’re all looking for new, exciting and useful ways for processing information to get our jobs done more efficiently. Does this sound like you?… “Please, just give me the information I need in a way I can understand. STOP making it so !@#$%^&en difficult and expensive! to get the information I need.” If this sounds familiar, read on.
This section (the main point I’m trying to make in this blog) shows several examples of how RIAs can make viewing, analyzing and understanding your data a lot easier-remember my tag-line “Map, Understand, Relate.”
We just saw an example of operational data (library locations) mapped on a Web map built with Adobe Flex. Now, lets look at census data.
Here are a few different ways of viewing the same data:
Perform ad hoc queries on a map
Using the same application from my GeoRSS blog http://www.roosdesignconsulting.com/dashboard this time, open the application by following this link and under the tools menu select demographics. When the widget tool opens use one of the drawing tools to draw a polygon anywhere on the map. A pie chart will open showing total population by age. The data is derived through a query on block group level census data, but you don’t actually see the block group layer boundaries, instead, they are in the background, the RIA mapping application just presents the query results in a pie chart.
To perform a more complex ad hoc query, follow this link http://www.roosdesignconsulting.com/dashboard.html and study example 1. This example shows an existing drive time service area defined using the service area widget under the tools menu. After you’ve explored this tool and generated your own service area use the select by polygon tool in the demographic widget to trace the outline of the drive time contour. The widget will return the total population by age for this area.
Another approach to census mapping using RIAs is to create a thematic map where you see the census data boundaries as seen in Figure 3. There are seven (7) census themes at the state, county and census tract levels contained in this application including: Median Household Income, Population Change 2000-2009, Population Density (per Sq Mile), Median Home Value, Unemployment Rate, Average Household Size, and Median Age.
The following example is, again, an Adobe Flex RIA, that show’s census data at the state, county and census tract levels. This is really cool stuff and I hope you are beginning to appreciate all of the ways RIAs can help you visualize data. Not long ago trying to get this type of information required a GIS professional and many, many hours of mapping and data design.
CLICK the image and the RIA map viewer seen below will open in a new Web browser.
Figure 3
Here are a few suggestions for navigating the map.
Map navigation using mouse:
- Press Shift and click map to recenter on clicked location
- Double-click map to recenter and zoom in at clicked location. To zoom out, press Alt when double-clicking
- Use mouse scroll forward to zoom in at current mouse location
- Use mouse scroll backward to zoom out at current mouse location
- Press Shift and click and drag to draw a rectangle around an area and zoom in. To zoom out, press Alt and Shift when drawing the rectangle
