PoliticsHome | Only the latest five entries on the PhiWire are visible to non-subscribers
- Sign up to see last 24 hours
Dont have an account?Sign up here
Thursday 14th April 2011 | 10:28
We reported yesterday on the Sun/YouGov tracker showing the "No" campaign breaking through a deadlocked race to establish a 7 point lead.
Now, a day is not usually a long time in tracking polls.
So today's AV IPPR/YouGov poll in the Independent showing a huge 12 point lead for the "Yes" campaign naturally raises some eyebrows. How could there be two radically different results on such a key issue from the same pollster?
Perhaps the answer lies in the way the questions were asked.
I've been passed the details of the questionnaire of today's poll, and they could not be more different.
Yesterday's poll, weighted for likelihood to vote, split the sample in two, asking each group just one question on AV. One group is asked the question wording on the ballot. Another group is asked YouGov's own wording explaning the how AV works and that the referendum was introduced by the 'Conservative-Liberal Democrat government'. The figures commentators give most weight to are the former sample, and this is what we commented on yesterday.
Today's poll asks 15 questions on AV, 14 of which discuss different voting systems and possible consequences - at the end of this volley, the poll asks AV voting intention question, as it appears on the ballot.
Among these 14 questions, are the following...
Which of these would you prefer?
A voting system in which extremist parties have a good chance of winning seats in a general or local election if they have the support of around one-third of local voters
A voting system which makes it very hard for extremist parties to win seats in a general or local election unless they have majority local support
Which of these would you prefer?
A voting system which encourages parties to have clear and distinct policies, and not to compromise
A voting system which encourages parties to reach out to voters as widely as possible, and to compromise when necessary
In the past have you ever engaged in 'tactical voting' in a general election - that is, voted for a party not because you liked it, or its candidate, best, but because you felt that your favourite party could not win in your area?
Yes, I have voted for my second choice party/candidate at least once
No, I have always voted for my favourite party/candidate
Do you agree or disagree with these statements:
“There would be less ‘tactical voting’ under AV, because people could give the party/candidate they like best their first preference, knowing that their second preference will count if their favourite candidate has only a little support.”
“I would be more likely to vote in an election held under AV because I would be able to express my preference for more than one party/candidate.”
Furthemore, the poll does not seem to be weighted by likelihood to vote (I am happy to be corrected on this), in an election which, as we discussed last week, will be determined by turnout.
This poll unquestionably excellent message testing for the "Yes" and contains some excellent research on the way in which voters perceive the debate. However, there are clear questions as to whether it can be seen as a real test of public opinion.
There is something rather Sir Humphrey about it all...