Countries contemplating a move to GIS for redistricting should give careful consideration to the advantages and disadvantages of introducing a GIS. The reasons for adopting such a system must be examined, and clear objectives identified for the use of the system. Questions that should be addressed include:
- What will be the benefits of introducing a GIS? Will it improve the efficiency or effectiveness of the redistricting process?
- What will be the costs of introducing a GIS for redistricting?
- Will the expected benefits outweigh the anticipated costs?
Potential users of GIS should conduct a cost-benefit analysis to determine whether the expected benefits of adopting GIS for redistricting exceed the anticipated costs of employing this technology.
Costs and Benefits of Using GIS for Redistricting
Some of the direct costs that may be associated with the use of GIS for redistricting are;
- Computer systems and software, including;
- hardware,
- software, software updates,
- customise software,
- maintenance and support of hardware, software.
- Data, including;
- electronic encoding of data,
- creating database,
- updating database.
- Users, including;
- hiring qualified personnel,
- training staff.
Some of the benefits that may be associated with the use of GIS for redistricting are;
- Savings, including;
- fewer staff required,
- time saving for routine and repetitive tasks.
- Increased effectiveness, including;
- faster provision of information,
- greater range of information provided,
- more accurate information provided.
- New Products, including;
- new range of output - maps, reports, etc.
- better quality output.
Some costs and benefits - such as the cost associated with purchasing the system and the benefit with regard to time savings - are relatively easy to identify and quantify. Other costs and benefits, however, especially indirect ones, are not as easy to identify and quantify but should not be neglected when doing a cost-benefit analysis
The cost of employing GIS for redistricting will vary dramatically depending on the availability of electronic data and the effort required to encode the data if it is not available electronically. If digitised maps of the entire country are available, the costs associated with adopting GIS will be much lower. If a digital map base must be created, and the cost of doing this cannot be shared with other government agencies or organisations, the effort required may not exceed the anticipated benefits of using GIS for redistricting purposes. If, however, a decision is made that the benefits do outweigh the costs, the next step in the process is to choose a GIS.
Choosing a GIS Software Package
Once the determination has been made to use GIS for redistricting, a decision must be made as to whether to develop the software in-house, purchase a standard GIS commercial package, or to purchase a standard GIS software package and customise it (using in-house staff or consultants).
Standard, proprietary GIS systems are likely to be cheaper, easy to use and well-supported but these systems may lack the some of the functionality required for redistricting in a specific country.
Some questions to consider when selecting a GIS software package are;
- What hardware is required to run the software? (What storage capacity, memory, speed, etc. is required to operate the software?)
- Which operating system does the software run on?
- Is the software user-friendly?
- What documentation is provided? In what languages is the documentation provided?
- What training is available?
- Will the software continue to be developed and supported by the vendor?
- Will upgrades to the software be provided?
- What functions does the software offer?
- Do the available functions meet requirements?
- Can the package be customised to meet requirements? If so, who is responsible for the customisation?
- How much does the software cost?
In addition to viewing demonstrations of GIS software and answering the kinds of questions listed above, a well-informed decision-making process might include benchmark tests and/or pilot projects with select GIS packages. For example, a benchmark test might be organised in which three or four GIS vendors are provided with a list of tasks that the potential buyer would like the GIS to be able to perform. The potential buyer might also supply data for use in the exercise. The vendors would then attempt to demonstrate that their system will meet the potential buyer's requirements. This exercise would require a fair level of expertise and awareness on the part of the potential buyer, particularly with respect to the identification of the tasks to be performed by the vendors. This is a good method of choosing between systems, however - far more effective than simply viewing vendors' demonstrations and discussing systems with vendors and other GIS users.
Implementing GIS for Redistricting
Once a GIS has been selected, it must be implemented. Even if there is nothing technically wrong with the GIS system chosen, problems may arise with its use depending on the how the system is implemented. Factors that could be important to the success or failure of the GIS implementation include:
Sufficient experience with and commitment to the use of information technology and digital data.
Sufficient training to enable staff to operate and support the system, as well as to encode the data (if necessary) and create a database.
Sufficient time and resources to create the database and experiment with the system before a final plan must be adopted.
The introduction of a GIS is not simply a matter of choosing the product, ordering the equipment and getting it running. It involves a complex interplay of technical and human factors. For this reason, it is important to carefully plan the implementation process and to make certain that there will be sufficient time and resources to ensure the successful transition from manual to computerised redistricting. If time or resources are lacking, it is quite likely that redistricting deadlines will not be met and that any plans produced will not be as accurate as possible. This, of course, will have major implications for the outcome of the process - the optimal redistricting plan may not be created or selected.
