Electronic voting in France increases counting errors

Polling stations using electronic voting systems suffered more voting discrepancies than polling stations using traditional paper votes in four recent French elections, according to a study sponsored by two groups campaigning for free and transparent elections.

Chantal Enguehard, a researcher at the University of Nantes specializing in electronic voting, looked at discrepancies between the number of electors who signed the electoral register to confirm that they voted and the number of votes subsequently counted for each polling station. The study compared discrepancies at 6,427 polling stations using electronic voting machines and 14,624 using paper ballots, in both rounds of the 2007 presidential election and two subsequent elections.

There were discrepancies between the number of signatures and the number of votes at around 29.8 percent of polling stations studied using electronic voting machines, compared to just 5.3 percent of those using paper ballots, and those discrepancies were larger in the stations using voting machines, Enguehard found. It’s unlikely that voters’ unfamiliarity with the machines is to blame, for two reasons, said Enguehard. The ratio of discrepancies between electronic and traditional stations got worse, rather than better, with time, and there was no correlation between the bureaus with discrepancies and the bureaus that received the most complaints about difficulties with the voting machines.

In French elections, voters are traditionally presented with an opaque envelope and a selection of ballot papers, each printed with the name of one of the candidates. In private, they place the ballot paper of their chosen candidate in the envelope, before identifying themselves to voting officials who verify that they are registered to vote and have not already voted in that election. Finally, they place their envelope in the transparent ballot box and sign the register of electors to say they have voted.

The opacity of the envelopes guarantees the secrecy of the vote, while the transparency of the ballot boxes guarantees that they were not stuffed with votes prior to the opening of the poll. Patient electors may observe the ballot box for the duration of the vote to satisfy themselves that the envelopes in the box are not tampered with, and that their number corresponds with the number of signatures in the register when the poll closes.

The introduction of electronic voting machines required a few changes to the rules and procedures. On arrival, voters identify themselves and their right to vote is verified. Then, they go to the polling booth where the polling station’s returning officer activates the machine for each voter. The voters select the name of their chosen candidate on the screen and confirm their vote before signing the electoral register and leaving. But there is no way for the average elector to verify the honesty of the voting machine as they can with a transparent ballot box, beyond the comparison Enguehard made of the number of signatures on the register and the number of votes recorded.

Enguehard’s study, funded by two campaign groups, Ethique Citoyenne, in Paris, and the European Computer and Communication Security Institute, in Brussels, concludes that a broader study is needed to determine whether the trends observed in this sample of polling stations hold true for the rest of the country. She also wants to look into the reasons behind the discrepancies.

Related content:

Interview with Aviel Rubin, e-voting activist and professor of computer science

E-voting the way forward for municipalities, survey finds

Related Download
Improving the State of Affairs With Analytics Sponsor: SAS
Improving the State of Affairs With Analytics
Download this case study-rich white paper to learn why data management and analytics are so crucial in the public sector, and how to put it to work in your organization.
Register Now