Scotland loses 150,000 votes
October 23rd, 2007 by Richard Carback in : Voting ProblemsYet another problem that could be found and/or avoided with the likes of Punchscan or Scantegrity:
The results process was besieged by delays and saw some 140,000 ballot papers rejected.
From the report:
The voter cast a valid vote on both the regional ballot paper and the constituency ballot paper for 96% of the Scottish parliamentary ballot papers counted. The remaining 4% of voters had one or both parts of their ballot papers rejected. The main categories for rejected ballot papers are:
• 50% of these voters cast a valid vote on the regional ballot paper, but left the constituency ballot paper unmarked (about 2% of all voters). In this case, the valid vote was accepted and the blank paper rejected;
• 25% of these voters cast a valid vote on the constituency ballot paper, but left the regional ballot paper unmarked (about 1% of all voters). Again, only the blank paper was rejected;
• Thus, 75% of these voters marked one cross only, on one or the other side of the combined parliamentary ballot papers (about 3% of all voters);
• Of the remaining 25% of rejected ballot papers, over half of voters ‘over-voted’ – casting two or more votes on the regional ballot paper (about 0.6% of all voters).For the 3% that voted for only one side the report offers three possible explanations including a deliberate choice to vote on only one side of the ballot paper; not understanding that there were two votes; and confusion caused by the appearance of named individuals on the regional list.
While this is inherently a usability problem, had they used Punchscan these voters would have gotten feedback from the scanner, rejecting over voted ballots, indicating what races had not been voted, and posting a receipt voters could use to check that the machine was honest online. DREs and other systems can also provide this kind of feedback, but there is no way to check that they are being honest with you.
