Estimated time:
Task |
Time (minutes) |
Introduction and preparation |
15 |
Manual interactive districting |
15 |
Automatic districting |
10 |
Finishing |
10 |
Total |
50 |
You will use GIS tools to do geographic districting exercise in this lab. The objective is to let you understand the geographic districting problem and teach you how to use GIS tools to find solutions to the problem. The sequence of lab exercise includes filling one pre-exercise and one post-exercise survey and using one manual interactive and one automatic GIS tool to carry out the districting task. This will be a timed lab, please follow instructor or TA's instructions to finish this lab.
You will need the following data and tools to finish this lab. Please do the followings.
distrinting.mxt (This is an ArcMap template containing the automatic districting tool.)
blockgrp shapefiles (These are the files associated with 47 census block groups in the NE Portland neighborhood.
Step I. Pre-exercise survey
Now, go to http://survey.oit.pdx.edu/ss/wsb.dll/jduh/districtingA.htm and complete the questionnaire.
Step II. Using ESRI districting extension
STOP +++++++++++++++++++++++++++++++++++++++++++++++++STOP
1) Select "Start Editing" from the Districting pull-down menu to begin the districting process.
2) Select "Preference..." from the Districting pull-down menu and check "Assign changes immediately after selection" option button. Click OK to dismiss the preference dialog window.
1)
Click on the Statistic Window icons to open the statistics window.
2)
Click the drop-down arrow after
"District:" on the Districting toolbar to select the district (i.e., D1, D2, or D3) to which the
selected polygon will be assigned. Select D1. Click
the Select Features icon and begin to assign
features to districts. Make sure individual districts must be formed by
contiguous block group polygons (i.e., the polygons must share edge segments). Select
as many block groups for D1 as you think will help to make an even
distribution. Then change the district to D2 using the District dropdown menu
on the Districting toolbar. Select what seems like a fair share of block groups
to District 2. Then assign block groups to District 3 in the same way. You can
also click and drag across block groups to select them, although you can't unselect
them, so be careful. You can change the district of a block group by selecting
the new district value from the District pulldown menu on the Districting
toolbar and clicking on that block group with the Districting Select Features
tool.
3) When done, click the Plan Report icon to show a report of the districting plan. Transcribe the POP90 values to the table below.
District |
Pop90 |
Pop90_dev |
1 |
|
|
2 |
|
|
3 |
|
|
|
Target Popu. |
Sum Absolute Diff. |
|
14977 |
|
STOP +++++++++++++++++++++++++++++++++++++++++++++++++STOP
Step III. Using the Districting Optimization tool
1) Click the Districting tool button to open the Automatic Districting dialog window. In the dialog window, click the browse button to select the "blockgrp.shp" file in the c:\temp folder.
2) After the blockgrp.shp map is displayed, set POP90 as the attribute field, change the # of zones to 3, make sure the "Use Contiguity Constraint" checkbox is checked (this will force districts to be contiguous). Accept all other defaults, and then click the Initialize button. Click Yes to overwrite the values in the districtid field. You can specify a different random seed number to generate different initial plans.
3) Click Run to start the automatic districting process. Click Stop to stop the process. After you click Stop, you will have to wait a little while to let the tool refresh the districting result in ArcMap. Be patient! A window saying "Optimization stopped" will appear. Click OK.
4) The default maximum number of trials is 5000. That means the optimization process will stop after 5000 trials. The best solution in these 5000 trials will be preserved and displayed in ArcMap when optimization stops. It takes less than 20 seconds to evaluate 5000 trials on the computers in the lab. You can initialize the optimization process with different random seeds (e.g., 5, 20, 389,...) and see which initial map results in the best solution. Just write down the random seed numbers and the value on the last row of the OFV (Objective Function Value) column in the report panel. The smaller the OFV value, the better the solution. You can always use the same seed number to regenerate an identical solution you got last time.
Random seed number used: ___________________________
District |
Pop90 |
Pop90_dev |
1 |
|
|
2 |
|
|
3 |
|
|
|
Target Popu. |
Sum Absolute Diff. |
|
14977 |
|
Step IV. Post-exercise survey
Go to http://survey.oit.pdx.edu/ss/wsb.dll/jduh/districtingB.htm and complete the questionnaire.
Please return this lab worksheet to the instructor before you leave the lab. Thanks!