National Study Finds Voters Don't Know Where to Vote on Election Day and Aren't Helped by Local Election Boards

facebook twitter google print email
Media Contact:Jacquie Posey | jposey@pobox.upenn.edu | 215-898-6460December 6, 2005

PHILADELPHIA -- The University of Pennsylvania's Fels Institute of Government today released two reports on the state of election administration in the U.S. that show a failure by local boards of election to answer voter questions on Election Day.  

The findings are detailed in the final reports on hotline data from the MYVOTE1 project, the largest-ever national election hotline conducted during the November 2004 U.S. presidential election and the 2005 elections in New Jersey, New York and Pennsylvania.

The national hotline, 1.866.MYVOTE1, is a partnership between the Fels Institute and InfoVoter Technologies Corporation.  

The 2004 national hotline, the first of its kind in the country, took in 208,524 calls, processed more than 102,200 poll-location requests, transferred more than 96,000 calls to local election officials and recorded more than 56,000 complaints across all 50 states during the November 2004 elections.

The 2005 hotline, operated in New York City, New Jersey and Southeastern Pennsylvania received 1,598 calls, processed 1,073 poll-location inquiries, recorded 278 complaints and attempted to transfer 830 calls to county boards of election.

The conclusions of the final reports on hotline data are that local boards of election do not answer their phones.

In 2004, nationwide, 47% of calls transferred to local election board voter hotlines went unanswered.  

In 2005, despite a 97% drop off in call traffic, local county boards phone systems failed to answer in-bound calls 30% of the time.

States with the most problems answering phones were Missouri, Pennsylvania, District of Columbia, New York and South Carolina. The reports conclude that the largest impediment to voting is lack of poll location information.

In 2004, nationwide, 51% of the hotline callers simply needed to find their polling places.

In 2005,  61% of callers simply needed to find their polling places.

States with the largest percentage of calls for poll locations on Election Day 2004 were Wisconsin, Texas, Indiana and New York. The reports concluded that registration and absentee ballot status comprise more than  50than 50% of the non-poll-location problems.

In 2004, nationwide, 55% of the 56,024 complaints recorded concerned registration or absentee-ballot systems.

In 2005, registration and absentee-ballot inquiries accounted for 56% of the non-poll-location call traffic.

Christopher Patusky, executive director of the Fels Institute, said that the MyVote1 data can be used to diagnose election-system problems much the way a doctor diagnoses disease based on evidence and not on guesswork or ideology.  According to Patusky, the data shows that passage of the Help America Vote Act may have solved the wrong problem.  "Congress distributed $3 billion to states, primarily for new voting machines, and they missed a far more pervasive and insidious problem: local governments are not making basic electoral information about poll location and registration easily available to the public, and this is likely disenfranchising millions of voters," he said.   

"Rather than use technology to create an information superhighway for voters, election officials have built an information cul-de-sac," Ken Smukler, creator of the national election hotline, said.  "By continuing to force voters into inadequate phone systems for basic information, we consign them to frustration, anger and, ultimately, disenfranchisement."

The project employed an Interactive Voice Response System developed by InfoVoter that could process thousands of calls from voters simultaneously.  The system offered voters an auto poll locator, the option to transfer to their local government's voter help lines and the opportunity to leave brief voice messages on the system.  The data from each transaction were stored in a database.  The voice messages were coded by sophisticated audio mining technology with results checked against a sample of 30% of the messages that had been human coded.  

Most municipalities and counties post their polling locations based on precinct numbers.  Since most people don't know their precinct numbers, they cannot learn their poll locations without calling local government help lines, and the MyVote1 data shows that many of these help lines are not reachable.

The MyVote1 project was sponsored by a bipartisan consortium of groups that included the Fels Institute, InfoVoter Technologies Corporation, the Reform Institute, the Common Cause Education Fund, the National Constitution Center and the Johns Hopkins University Hispanic Voter Project.  The project was funded by the Carnegie Corporation, the Open Society Institute, the JEHT Foundation and the John S. and James L. Knight Foundation.   

The full Fels Institute Report with an accompanying InfoVoter data appendix can be found at www.fels.upenn.edu/research_service.htm.