YouGov poll suggests that the pro-independence campaign's childcare proposals have made a big impression on voters

The remaining details of this week's YouGov poll have been released, and the most eye-catching finding is that there are now more than twice as many voters (35%) who think that an independent Scotland would have better childcare provision than there are who think provision would get worse (15%). Even more remarkably, a plurality of Labour voters (27% to 18%) think that independence would lead to expanded childcare provision, while as many as 15% of those who are currently planning to vote No in the referendum share the same view - meaning that, by a fair distance, childcare is the issue that provokes the most favourable impression of independence among current No supporters. That leaves little room for doubt that the decision to make childcare the centrepiece of the independence White Paper has been noticed by many key voters, and that the No campaign's attempts to neutralise the impact of the proposals have thus far been unsuccessful.

One of the subtexts of John Curtice's commentary on recent polls is that, because Yes have failed to storm into an outright lead on the back of the childcare proposals, the strategy must have been wrong, and that they should have focused relentlessly on the economy instead. I disagree. The economy is undoubtedly a vital battlefield, but the idea that there was some kind of silver bullet available that would have produced an overnight 10% swing in voting intentions is risible, and the idea that dry talk about economic growth or tax revenues could have constituted that silver bullet is even more risible. No, the task of the White Paper was not to instantly shift votes in huge numbers (although as we've seen in the polls that have been published since then, some voters have certainly moved over to the Yes camp), but instead to shift perceptions of what independence is all about among the most sceptical voters, many of whom are women, and many of whom care deeply about issues such as childcare. With a change in perceptions you earn a fair hearing further down the line that you might not otherwise have got, and with a fair hearing you have the chance to ultimately win new votes - if your campaigning is skillful enough, that is. There's every indication in these YouGov numbers that things are going to plan so far.

The other thing that struck me is just how few issues there are that really provoke any fear about independence, even amongst people who are currently planning to vote No. For example...

* 53% of current No supporters think that schools would be as good as now or better after independence.

* 62% of current No supporters think that the crime rate would not get worse or would reduce after independence.

* 45% of current No voters think that Scotland would be just as 'safe in the world' (whatever that means) or safer after independence. (This also constitutes a plurality, because only 42% of No supporters believe Scotland would be less safe.)

So we're not exactly dealing with voters who have an all-encompassing dystopian view of independence - their reluctance to embrace the idea seems to boil down to a relatively narrow (albeit very important) range of concerns about issues such as taxation and pensions. The poll also detects a nominal 'concern' about Scotland's influence in the world, which of course is entirely misguided - Scotland can hardly lose any further influence in world councils when the people 'representing' us there at the moment are David Cameron and William Hague. But I don't think it really matters whether the electorate come to accept that fact or not, because the only people who ever actually lose any sleep over a declining influence in the world are to be found in the political class.

Now here's an interesting paradox for you. The poll shows that most people in Scotland think of themselves as having either a wholly or predominantly Scottish national identity (ie. they say they are either 'Scottish not British' or 'more Scottish than British'). The poll also shows that most people with a wholly or predominantly Scottish national identity are in favour of independence. So why aren't Yes already in the lead? Well, quite simply because the No vote encompasses a substantial minority of those with a mostly Scottish identity, and a big majority of those without one. But all the same, this represents a huge potential opportunity for Yes due to the power of example - if voters who think of themselves as primarily Scottish start to notice that most people of like mind are plumping for independence, there might well be a bandwagon effect. There are no guarantees, of course - I mentioned Quebec last night, and the pro-independence campaign there were stuck with their own paradox of losing narrowly in spite of a 60-40 split in their favour among the majority French-speaking population. But the possibility is certainly there.

It's also intriguing that more than twice as many respondents (10%) say they are 'British not Scottish' than opt for the more nuanced option of 'more British than Scottish' (4%). It seems to me the most plausible explanation is that most of the 'British not Scottish' group are literally non-Scots - they're people resident in Scotland who have come here from other parts of the UK. By and large, that means they'll be English people answering a question they feel doesn't really apply to them in the only logical way they can - that would certainly explain the counter-intuitive finding that a respectable minority of them are planning to vote Yes or are SNP supporters. If this theory is correct, then it suggests that Scots with a primarily British national identity are now almost an extinct group, and that the No campaign's hopes depend heavily on retaining support among the quarter of the electorate who feel 'equally Scottish and British'.

Lastly, the poll backs up the suggestion from Ipsos-Mori that a swing in favour of independence has occurred in spite of a simultaneous small swing to Labour in Holyrood voting intentions. That suggests that referendum voting intentions are becoming gradually decoupled from party loyalty - and that's a very good thing, even if a modest boost for Labour in the polls isn't. The SNP do retain a slender lead on the constituency vote among respondents who are 100% certain to vote, which many pollsters regard as the most accurate test of how the electorate would actually vote at any given moment.