Digesting and assembling TIGER files for a regional topological study of demographic boundaries

By Paul Cote
For Guy Stuart
March 21, 2003

Overview: This document describes a procedure for digesting and assembling census TIGER files for a regional study examining the types of boundaries that divide demographic areas in a region.

Introduction

1. Examine the sample datasets and their directory structure
2. Look at the final product in ArcGIS

Running the Program

1. Install ArcGIS and ArcInfo on a computer running windows NT, 2000 or XP.
2. Running the program with the sample data.
3. Converting results to geodatabase format
4. Getting TIGER Files for other areas

Examine the sample data sets and their directory structure

This set of procedures is intended to take raw tiger files, to convert them to ArcInfo coverages, do various manipulations on these, join them with a table of demographic variables. The ouput is an ArcInfo coverage database that and then put the result into a ArcGIS geodatabase, with a bunch of standard layers so that they can be easily explored in ArcMap.

The directory on the CD, named asmple is an example of a directory that would be created for any geographic refion that would be studied. This directory contains the raw material, the scripts and the outputs related to the procedure as applied to Suffolk and Norfolk counties, Massachusetts.

The files and directories in sample:

• Raw TIGER files are kept in the directories 25025 and 25021. The program requires that these directories be named with their fips state and county codes.
• Arc Macro Language Scripts the files Calvin.aml is the main program for digesting tiger files. lpad.aml is a program that inserts the apropriate amount of 0 characters in front of short tract IDs, which are handled as integers in ArcInfo. Tmp.aml is a program that is created on the fly by calvin.aml.
• The directories, tgr250251, tgr250253 tgr250211 and tgr250213 are the intermediate polygon and point coverages that are created by calvin. The directory info contains information related to each of these coverages. These are kept around to save time in case the program needs to be run again, say to add another county to the region.
• The file, geolitics.dbf is a dbf file of demographic attributes. It must contain the following columns:
  • arealand the land area of each block in meters
  • segindex some index of segregation for each block. in this example, it is simply, percent white.
  • racepop the apropriate universe population for the race information in each block
• The directory arcmap contains the final output of calvin.aml, the arcinfo coverage, blocks and its related info directory. into which all of the block polygons and arcs have been exported. The file, cfcc_lut.dbf is a lookup table for Census feature Class Codes. All of the files named *.lyr are arcmap layers that deal with symbolization of various attributes of the polygons and arcs. The file arcmap.mxd is an ArcMap map document that previews the output in various symbolizations

Look at the final product in ArcGIS

If you use ArcGIS to open the file arcmap\arcgis.mxd you will be able to look at the geometric results, and the associated tabular attributes.

The layers in arcmap.mxd are different views of the polygon and arc feature classes that resulted from the CALVIN procedure. These are linked with the census feature class lookup table, or the table of block demographic attributes. Roll over the names in the left column of the following table to see illustrations.

Water Polygons
Elementary and Unified School Districts
Voting Districts
Polygon Landmarks
Key Geographic Locations
Percent White
Arc (Line) Features

The topological structure of TIGER files and ArcInfo coverages stores the relationship between each line segment and the polygons that fall on the left and right side. Which side is the left, and which is the right is arbitrary. If we zoom in on the map and use the info tool, we can look at the attributes of the polygons and lines. For example, the cfcc attribute for each line designates the type of line, which may be a physical or invisible attribute. The attribute for segdif contains the difference in the segregation for each side of the line. The attribute, length, tells us the length in meters for each of the lines. This table should give us enough information to answer questions such as: What is the total length of arcs that are voting district boundaries? What is the total length of arcs that are voting district boundaries AND divide blocks having a difference in segregation over a certain amount?

Arcs by Census Feature Class Code
Arcs that are voting district boundaries
Arcs colored by the differencr in Segregation on either side
An arc showing its left and right attributes.

The complete attribute tables for each layer in ArcMap may be accessed by right-clicking on the layer name and selectin Open Attribute Table. Once opened, the attribute table may be exported to a dbf file by selecting Export from the options menu at the bottom of the table.

Here is a data dictionary for the arc attribute table.

Running the Program

To run this program, follow these steps:

1. Install arcgis desktop (complete installation, no need to install crystal reports or the tutorial data.)
2. Install ArcInfo Workstation
3. Copy the entire sample data directory to your hard drive You may name it according to the study area you plan to digest (e.g. c:\chicago). Be sure to change the properties for this new folder, unchecking the read-omly attribute for the folder and all of its contents.
4. Put unzipped TIGER files into directories named for their statecounty fips codes (e.g. suffolk county is 25025)
5. Open the file, calvin.aml, and edit the user editable parameters as indicated at the top of the file.
6. Start the Arc program from the ArcGIS/Workstation Arcinfo program group.
7. At the arc propmt, set youe workspace to your new directory

   arc: w c:\chicago
   

8. Run the script calvin.aml

   arc: &r calvin.aml
   

   If you have run calvin before, there may be pre-digested TIGER file coverages existing. Calvin will ask if you want to remove these. If the last run of calvin was successful, then you should keep them. This will save a bunch of time. Specifying 'all' at this prompt will remove them all.
9. When finished, you should be able to open the arcgis.mxd file from the arcmap directory to view the results.