An Opportunity to Make Every Vote Count
Manual placement of voters in districts is a recipe for error. The solution is to enhance the use of GIS in the process.
Last year a Virginia House of Delegates race drew headlines across the country -- and it had little to do with the candidates and their campaigns. When the votes were tallied, there was a tie. The race ended with a judicial panel drawing a name out of a hat, thereby determining party control of the House.
But that wasn't the end of the story of the District 94 race. Earlier this year, the Washington Post reported that 26 voters in that pivotal district had been misassigned. Due to human error, citizens who went to the polls were given the wrong ballot. In another Virginia race, for House District 28, at least 147 voters were assigned to the wrong district, and the margin of victory there was 82 votes.
While we'll never know who would have won those races had voters been provided the proper ballots, these examples serve as a reminder that details matter in every election. Placing voters in their appropriate precincts, districts and wards is a critical part of election administration, and errors are not limited to Virginia. Reports of misassigned voters in Ohio and Georgia surfaced in this year's primary election cycle, and in South Carolina in a special election. This can, and likely does, happen all across the country.
The fact is that when it comes to election administration, no margin of error is acceptable. But no matter how talented, efficient or passionate election officials are, human error is always possible. A labor-intensive, manual process escalates the possibility that mistakes can be made and missed.
The good news is that there are solutions. Geographic information systems provide a more accurate way not only to draw district lines but also to place voters within their proper voting precincts and districts. Most of the governing bodies drawing district lines do so using GIS, but then in most cases share the converted district information to election officials in textual representations, called metes and bounds, which require the manual placement of voters. That often is the cause of these later-discovered inaccuracies.
There are states and other jurisdictions, however, that are enhancing their use of GIS for more accurate placement of voters. A Democracy Fund Voice grantee, the National States Geographic Information Council, recently released its State Representatives Baseline Survey Report as part of its ongoing project to help disseminate best practices and tools to aid in the expansion of GIS use in elections. Seventy-nine percent of the survey's respondents reported that their mapping of state-level district-based elected offices is now accessible online in a GIS format.
It is a missed opportunity, and one that should be remedied, to have GIS in use in some facets of state government, such as transportation and environmental protection as well as the initial drawing of voting-district lines, but not have it available for election officials' use for the placement of voters and conducting elections. Once the 2020 census is complete, states will embark on the redrawing of lines and districts that will result in the movement of voters. This is a junction at which we must decide whether to modernize our method of placing voters across jurisdictions -- as Utah, Wisconsin, Los Angeles County, Calif., and Maricopa County, Ariz., already have --to create a more accurate and efficient GIS-driven model, one that is nimble enough to adapt when district and precinct lines change.
The Democracy Fund supports the creation of this advanced form of election process modeling of districting and placement of voters and has helped launch a technical workbook detailing these processes. We have time before the next round of redistricting to put the proper infrastructure and statutory authority in place to mitigate manual-entry errors like those that occurred last year in Virginia. And there is an additional benefit to these improvements: A more accurate process may uncover old errors. In Fluvanna County, Va., for example, when GIS was used to confirm county borders officials found that a couple had been assigned to, and voting in, the wrong district for almost 40 years.
While it is hard to tell if implementing these improvements would change the outcomes of elections, it would identify existing errors and prevent new ones while allowing for a better and more efficient system and enhancing citizens' confidence that their votes truly count. If we can implement these changes before the next round of redistricting -- and I'm confident that we can -- it will ensure that the United States' voting system is one that is better for all of its citizens.