It’s now exactly one month to go until the Scottish referendum, and this morning’s Times has a fresh YouGov Scottish poll. Topline figures are YES 38%(+3), NO 51%(-4). Excluding don’t knows this works out at YES 43%(+4), NO 57%(-4). The 43% YES figure excluding don’t knows is the highest YouGov have shown so far.

The previous YouGov poll straddled the Salmond-Darling debate, so this is their first test of support post-debate. Looking at all four of the post-debate polls, we’ve now got Survation showing a significant move to NO, Panelbase and ICM showing modest movement towards YES and YouGov showing a significant shift towards YES.

Clear as mud, but I think it’s fair to conclude that despite Alistair Darling emerging as the initial “winner” of the first debate, the broad trend amongst the post-debate polls is looking like things may have actually moved a little in the YES direction.

Looking back at the post-debate poll, it might be worth remembering that existing NO supporters tended to think Darling won, existing YES supporters tended to think Salmond won – so Darling’s “victory” was largely a result of him having more supporters to begin win. If subsequent polls do confirm that there has been a movement to YES since the debate, perhaps we’ll conclude that attitudes towards who won the debate were different amongst swing voters…


There are two new Scottish polls in Sunday’s papers, an ICM poll for the Scotland on Sunday and a Panelbase poll commissioned by the Yes campaign. These are only the second and third polls that we’ve seen conducted wholly after the debate between Alex Salmond and Alistair Darling, the first one from Survation suggested a significant move towards NO, but these two paint a different picture.

Taking the Panelbase/Yes Scotland poll first, topline figures are YES 42%(+1), NO 46%(-2), Don’t know 12%(+1). Excluding don’t knows this works out at YES 48%(+2), NO 52%(-2). Changes are from the Panelbase/Sunday Times poll in July. Compared to last month it’s a slight move towards YES, but is not particularly significant by Panelbase’s standards: their penultimate poll also had YES on 48%, and they’ve have YES on between 46-48% since March.

The ICM poll in the Scotland on Sunday has topline figures of YES 38%(+4), NO 47%(+2), Don’t know 14%(-7). Excluding don’t knows this works out at YES 45%(+2), NO 55%(-2). It’s a bigger NO lead than Panelbase are showing, but the same modest movement towards YES. As with Panelbase, it’s not a massive change from the longer term trend – so far this year ICM have had YES between 43% and 48%, and YES 45% is right in line with the average of all ICM’s Scottish polls this year.

So, two polls, both show a modest movement towards YES since last month, but neither a significant shift from the longer term trend. What it does mean though is that the movement towards NO in Survations’s post-debate poll has not been echoed in other companies’ polls.

(Incidentally, I’m on leave this week, having a break before the long slog to the general election, so expect light blogging for the next few days)


Kevin Schofield at the Sun has just tweeted out the latest YouGov Scottish polling from tomorrow morning’s Sun (£). Topline figures are YES 35%, NO 55%. Without don’t knows it works out at YES 39%, NO 61%. The fieldwork for the poll straddled the debate – just over half took place pre-debate.

39/61 is exactly the same as the last Scottish YouGov poll, but it isn’t directly comparable. There are two slight changes in YouGov’s Scottish methodology since the previous poll. The first is that the sample is extended to include 16 and 17 year olds – though this didn’t actually make any difference to the result.

The second is that YouGov have added an extra weighting variable, weighting according to people’s country of birth. For some reason raw samples seem to contain too many respondents who were born in England, and English born people are more likely to vote NO (Panelbase found the same, and also adopted place of birth as an extra weighting variable in their latest poll). This additional weight does makes a slight difference to final result, making the results slightly more “YES”. Under the old weighting scheme the results would have been YES 38%, NO 62%, a slight shift towards NO.


This morning’s Scottish Daily Mail has a new Survation poll of Scotland. Referendum voting intention figures with changes from last week are YES 37%(-3), NO 50%(+4), Don’t know 13%(-1). Excluding don’t knows that works out at YES 43%(-4), NO 57%(+4). Tabs are here.

Survation are one of the companies that have tended to produce some of the better figures for YES. Their last three Scottish polls have shown figures of YES 47%, NO 53%. 43% is the YES campaign’s lowest score in a Survation poll since January, before Survation switched their weighting scheme to Holyrood recalled vote.

Normal caveats about any sharp movement in the polls apply, but this the first full natrep Scottish poll since the Salmond-v-Darling debate, so obviously people will look at that four point swing and conclude that the debate has indeed moved public opinion towards NO. I’d urge a little caution – as ever, it is just one poll, all polls have a margin of error, and when other post-debate polls come along they may or may not paint the same picture. This first bit of evidence though suggests the first debate has helped NO.


Tonight is the long awaited Scottish debate between Alex Salmond and Alistair Darling. STV released their latest Ipsos MORI at the start of the debate – topline figures there are YES 40%(+4), NO 54%(nc), don’t knows just 6%. Excluding don’t knows that works out at YES 42%(+2), NO 58%(-2).

MORI tend to be one of those pollsters who show more favourable figures for the NO campaign, so by their standards its a favourable poll for YES. Then again, if MORI are right, then a sixteen point lead for NO is still a a big gap to close with only six weeks to go.

Following the debate the only instant poll I’m aware of is ICM for the Guardian, due to go out about 9.40 (results will hopefully be before ten, but it obviously depends on how quickly people respond!)

UPDATE: ICM’s instant poll crowns Darling the winner – 56% for Darling, 44% for Alex Salmond. The figures are, incidentally, very close to the sort of NO/YES figures ICM report in referendum voting intentions. We’ll know properly when we see ICM’s tables, but I suspect we may find that people who were voting YES anyway thought Salmond won, people who were voting NO anyway thought Darling won.

UPDATE2: Full figures including don’t knows were Darling 47%, Salmond 37%, Don’t Know 15%. Sample size was 512.

UPDATE3: Tabs are here. People’s perceptions of who won were, as suspected, largely in line with their pre-existing dispositions towards independence, though not entirely. Amongst people who were voting NO before the debate people thought Darling won by 83% to 6%. Amongst pre-debate YES voters people thought Salmond won by 72% to 16%. Amongst people who said they were don’t knows, Salmond was slightly ahead – 44% to 36% (albeit, there were only 63 don’t knows, so we’re talking about the difference of 4 or 5 people). Bottom line is that there was no big knockout blow here – the large majority of both sides thought their own “champion” won, don’t knows were pretty evenly split.