DRS Home
DRS News
◄ Back to DRS news

Sep 04, 2007

No question over e-Counting accuracy in Scotland

DRS e-Counting technology is designed to accurately process the most complex elections using a combination of voting systems including first past the post, additional member systems and the extremely complex single transferable vote. DRS technology has proven to improve the accuracy of results, increase security, reduce the human administrative burden and deliver reliable, robust and accurate results.

As world leaders in scanning excellence, DRS technologies have been used all around the world to create accurate voter registration lists, process complex election counts and conduct national population census projects. The technology is highly accurate and is tailored to the unique validation and security rules as defined by the governments and clients employing it.

In response to last night’s BBC Reporting Scotland and Newsnight Scotland reports, which called the accuracy of the e-Counting technology deployed across the Scottish Parliamentary and Local Government elections in May 2007 into question, DRS releases the following statement:

Every election requires validation or election rules. These rules are defined not by DRS, but by the governments or organisations undertaking the election. The validation rules applied to the e-Counting systems deployed in Scotland to identify when a ballot paper should be recorded automatically, or passed for further scrutiny, were defined by the e-Counting Project Board, on which the Scotland Office sat alongside the Scottish Executive, representatives of the 32 Returning Officers for Scotland, and of course the contractor – DRS.

These validation rules were applied to the e-Counting system by DRS, as directed, and employed consistently across every count in Scotland.

The e-Counting machines applied a whole series of complex validation rules which were defined for each agreed ballot paper scenario. All ballot papers were subjected to a high level of security checks. Depending on these validation rules, some ballot papers which surpassed all security checks and validation checks were read or adjudicated automatically, and some were sent on for further scrutiny by Returning Officers and their staff.

Once the e-Counting system was set-up to operate under these rules, a thorough testing programme was undertaken. During this testing, ballot papers of all scenarios – including blanks and partial blanks - were fed through the e-Counting system to ensure that the way in which this system recorded the votes, or passed the ballot papers for further scrutiny, met the specification as supplied to DRS. This specification was documented and the scenarios when auto-adjudication should take place were clearly defined.

The group that undertook this focused, and thorough, testing included representatives from the Scotland Office, Scottish Executive, AEA, SOLAR and SOLACE.

This same method of auto-adjudication of blank ballot papers was used at every testing stage in the lead up to the election, including the trial stage, acceptance stage and the large scale volume test that took place in Edinburgh in October 2006. These stages were defined by the e-Counting Project Board and were attended and signed off by representatives of both the Scotland Office and the Scottish Executive. These events were also attended by representatives of all 32 Scottish Authorities and many representatives of Political Parties and the media.

The use of auto-adjudication was contained in the contracts that were signed by all stakeholders in the combined election project – DRS, the Scottish Executive, the Scotland Office and the 32 Returning Officers.

There is no question as to the accuracy of the technology. Any ballot paper where there was any doubt over the voter's marks was sent for manual adjudication.