It's actually been quite difficult to get to grips with the new Survation poll that has been gradually released by Cleggy and the Vow-Meisters over the last few days, because as far I can see only the datasets for the questions on independence have been published so far. Survation's website isn't the most user-friendly, though, so it's difficult to be 100% sure. The latest figures to be released today relate to the removal of powers from the Scottish Parliament, and on the face of it would seem to confirm that the issue has now cut through - in spite of the heroic efforts by the BBC and other parts of the mainstream media over the last few months to mention it as little as possible and to downplay its importance whenever they do mention it. 41% of respondents agree that Westminster grabbing back 24 of Holyrood's existing powers amounts to a "power-grab" (logical enough, you'd think), while 34% disagree. Perhaps the closeness of that result may seem a little disappointing, but when you consider that the SNP have been fighting against a virtual news blackout on this subject, I'd suggest we should look upon the glass as being very much half-full in this case.
The only caveat I'd add is that in the absence of the datasets it's not clear exactly what question Survation asked. As we all know, people are very hostile towards the Tories and suspicious of the UK government's intentions (with good reason), and so it could be that if they were asked "the Tories are doing X, but say it doesn't mean anything, do you believe them?", that could have generated the 41% result without there necessarily having been as much pre-knowledge of the power-grab as we'd like to believe. But we'll find out more when the wording of the question is eventually published.
I've been a bit tied up over recent days writing articles for The National and iScot (and yeah, OK, watching the World Cup and Wimbledon may have had something to do with it as well), so I didn't get round to adding some analysis of Survation's voting intention numbers for Holyrood and Westminster. Here are a few belated thoughts. I speculated in my piece in The National that the SNP's best poll showing since before the June 2017 election might be due to the walkout from the Commons a few weeks ago. Of course the other potential game-changer was the Chequers "deal" and the subsequent spate of ministerial resignations, which took place in the middle of Survation's fieldwork period. We should certainly take that seriously as a possible explanation, because there's plenty of polling evidence that it's shifted public opinion at Britain-wide level - there's been a swing from Tory to Labour that essentially reverses the trend of the last few months. The odd thing, though, is that the Scottish Tory vote is not substantially down in the Survation poll - they've remained static in the Westminster vote, and have only dropped one point on the Scottish Parliament constituency ballot, which can easily be explained away as margin of error 'noise'. (They're down four points on the list ballot, but I'd be inclined to take that less seriously given the apparent tendency of some respondents to treat the list as a second preference vote.) Weirdly, it's Labour that appears to be suffering the most - the opposite of what is happening at Britain-wide level. How do we explain that? Perhaps pro-European voters are looking for the best available option, and in England that's Labour, but in Scotland it's the SNP?
Both UKIP and the Liberal Democrats - parties that are on opposite extremes of the Brexit debate - are doing unusually well in the Survation poll, which would tend to confirm that Europe is on voters' minds to a greater extent than usual, and might suggest that other changes in the poll have a similar explanation. In case you've been wondering why the seats projection for Holyrood gave the pro-independence parties a majority of seats on a minority of the vote, part of the explanation is that UKIP's 5% list vote is effectively 'wasting' a significant chunk of unionist votes, because it's just short of what would be required to actually win any seats. If UKIP's list vote was to continue to rise, or if all the UKIP votes were to go to the Tories, the seats projection would look somewhat less favourable from a pro-independence point of view.
In relation to Survation's inexplicable decision to suddenly stop including 16 and 17 year olds in their independence polling, someone asked on an earlier thread whether that meant they were also excluding EU nationals. The answer is that I don't know, because that information simply isn't available in the datasets. If Survation are now using the Westminster franchise rather than the Holyrood/local government/indyref franchise as the basis for their sampling, it would seem logical that they would be excluding EU nationals as well as 16 and 17 year olds, which might theoretically be leading to a marginal underestimation of the Yes vote. But that's just speculation at this stage.