Saturday, May 05, 2007

Scotland's butterfly ballot

Much attention has been given, rightly, to the confusion that has been caused by holding elections under two different PR systems (one, a "mixed" system already) on the same day.

While this confusion clearly did exist, and many voters put numbers or Xs on the wrong ballot papers, most of the spoilt parliamentary papers I saw were not like this. A great many had two Xs on the list vote, and nothing on the constituency vote. And many had a single vote on the list, and nothing on the constitency - and so were valid votes on the list.

Why were the constituency and list votes on the same ballot paper at all? Because, I suppose, this makes the scanning process much more efficient. Obviously with hand-counting, two contests on one ballot paper would be more work, but with machine scanning, it is less.

The list contest was on the left half of the paper. This makes it the 'first' vote. The greens, for example, after successfully winning seats with the message 'second vote green' have had to campaign 'first vote green' this time. This is an acceptable strategy, but it is bound to lead to confusion between first vote and second vote, and first choice and second choice - and hence spoilage rates. The psychological advantage the greens had previously, that they may get second votes from people who considered them a second choice, has gone. This will have harmed them, but that was never an advantage they were entitled to keep. On the other hand, a lot of papers spoiled with two votes on the list included one for the greens or socialists, and so they do have a grievance there.

What must have aggravated the confusion was the way that the list options took the entire column on the left of the ballot paper, and the constituency candidates were at the top of the right column. This looks very much like one long list carried over into a second column. How many voters saw it thus, and thought they had one, or two votes to be used anywhere in the long list? Tens of thousands, it seems, that's how many.


On the question of the computer failures, I am looking forward to the conclusions from the enquiry. But I am rather less concerned here than I am about some of the systems used in England, that are completely paperless, and thus have no way for results to be checked, and, frankly, for which we have no grounds to trust the result. While scanning does have transparency issues, these are an order of magnitude less than they are for things like internet voting. Yes, it seems a fix had to be applied to a live election system, and this itself can be a security question mark. I hope the electoral commission has the know-how to ask the right questions here.

2 comments:

James Graham (Quaequam Blog!) said...

From what I was able to make out, most of the problems with the voting machines in Scotland were rooted in the fact that the audit trail didn't match the result. That would appear to be a case of the software failing, but the system working, which is actually fairly reassuring.

Joe Otten said...

James,

Have you a source for that? I would like to get to the bottom of this.