The full datasets from the YouGov poll are now out, and they reveal a treasure-trove of information from supplementary questions. Most significantly, the exercise that Panelbase undertook a few weeks ago has been repeated, with respondents asked to say whether they think each of a range of administrative/legislative powers should remain reserved to Westminster or be devolved to the Scottish Parliament. In broad terms, the results are identical to Panelbase, with a clear desire for maximum devolution (or "Devo SUPER Max" as was so memorably promised by the No campaign's official representative in the TV debate at the Hydro). By a margin of 67% to 25%, voters want income tax powers to be devolved to Edinburgh, by 67% to 24% they want other taxes like corporation tax to be devolved, by an astonishing 71% to 22% they want working age benefits such as Jobseekers' Allowance to be devolved, by 47% to 45% they want the state pension to be devolved, by 60% to 33% they want the minimum wage to be devolved, by 60% to 32% they want health and safety regulations, consumer protection and competition law to be devolved, by an eye-popping 86% to 8% they want the running of Scottish elections to be devolved, and by 50% to 40% they want laws relating to abortion, embryo research and medicine to be devolved.
The only exception is that there is very narrow support - by 45% to 43% - for "broadcasting regulation and the BBC" to remain reserved to Westminster. That contradicts the Panelbase finding, and in my view it's somewhat misleading due to the question wording. By specifically bringing the "British Broadcasting Corporation" into it, some respondents are likely to be puzzled as to how a Britain-wide organisation based in London can be devolved to Scotland. My strong guess is that if the question had simply asked about broadcasting regulation, there would have been a similar result to the Panelbase poll, with a narrow majority in favour of devolving the powers to the Scottish Parliament. Having said that, the fact that the result is essentially level-pegging even when respondents are confronted with the sacred cow of the BBC is quite telling.
The other key finding from the poll is nothing short of devastating for Jim Murphy and the Labour party. Voters are asked how they would vote in the 2016 Holyrood elections on the hypothetical basis that Nicola Sturgeon is the SNP leader and Jim Murphy is the Labour leader - and the answer is no different to the headline Holyrood results, with the SNP maintaining a whopping 18% lead. Normally when a party is looking to be saved by a "king over the water" (such as Michael Heseltine in 1990), hypothetical "named leader" polls produce radically different results, so quite plainly voters do not regard Murphy as the game-changer that London bubble journalists have convinced themselves he is. And it can't really be credibly argued that voters will be more impressed by Murphy when they see more of him - he's barely been off the bloody television for months. By contrast, the SNP lead is reduced (albeit only to 13%) when voters are pointlessly asked to imagine that Gordon Brown is the Labour leader, although even then the SNP vote isn't affected - Labour's extra support is drawn from other parties.
* * *
SCOT GOES POP POLL OF POLLS
This is the first Poll of Polls update we've had that has included two full-scale Scottish polls (from Ipsos-Mori and YouGov respectively), so it's the most credible one so far. It also factors in five Scottish subsamples from GB-wide polls - four from YouGov and one from Populus.
Scottish voting intentions for the May 2015 UK general election :
SNP 45.8% (-1.4)
Labour 25.6% (+1.2)
Conservatives 13.8% (+0.9)
Liberal Democrats 5.0% (-1.1)
Greens 4.7% (-0.4)
UKIP 4.1% (+1.0)
(The Poll of Polls uses the Scottish subsamples from all GB-wide polls that have been conducted entirely within the last seven days and for which datasets have been provided, and also all full-scale Scottish polls that have been conducted at least partly within the last seven days. Full-scale polls are given ten times the weighting of subsamples.)