For a Thursday there was rather a lot of polling today which I’m only just getting chance to catch up with.

Firstly we had Ipsos MORI‘s monthly political monitor for the Standard. Topline voting intention figures are CON 31%(-1), LAB 37%(+2), LDEM 9%(-4), UKIP 15%(+4) – a good boost for UKIP following the Clegg-Farage debate. The 15% for UKIP matches the highest they’ve ever received from the pollster, last reached in April 2013.

The rest of the MORI poll had some questions on perceptions of the leaders, which showed the familiar comparisons between Ed Miliband and David Cameron: Cameron is seen as a more capable Prime Minister and better in a crisis, Miliband is seen as less out of touch. MORI also found a budget bounce in George Osborne’s reputation, nudging his approval rating into positive territory. 47% are now satisfied with his performance as Chancellor, 44% disatisfied, the best MORI have found for a Chancellor since 2006 (and the best for a Tory Chancellor since 1980). Full details of the MORI polling are here.

The second GB poll of the day was the daily YouGov poll for the Sun. They had topline figures of CON 33%, LAB 36%, LDEM 10%, UKIP 14%, but also had some Maria Miller questions here. 83% think she was right to resign, 63% think Cameron should have sacked her immediately rather than standing by her.

Moving on from GB polls, there were also two Scottish referendum polls, one showing a slight but insignificant drop for YES, one showing things static. The first by Survation for the Record had referendum voting intentions of YES 37%(-2), NO 47%(-1). Tabs are here. Interestingly enough Survation also asked Scottish voting intentions for the European elections. Most Scottish voting intention questions at the moment don’t interest me that much given the referendum result will shake things up either way, but the European election obviously comes before the referendum. Survation have figures of CON 13%, LAB 30%, LD 6%, SNP 39%, UKIP 7%. That would give the SNP three MEPs, Labour two and the Conservatives one. The Lib Dems would lose theirs and UKIP would fail to break through in Scotland.

The second poll was a Panelbase one commissioned by the YES campaign, which showed the same five point lead for NO recorded in the previous two Panelbase polls: YES 40%(-1), NO 45%(-1). Tabs are here.


Just catching up on another Scottish poll at the weekend – Panelbase produced a new referendum poll for Wings Over Scotland. Topline figures are YES 41%(+1), NO 46%(+1), confirming the narrower lead we saw in the last Panelbase poll in mid-March. Full tabs are here.

There remains a considerable contrast between pollsters on the Scottish referendum so it’s important to compare like to like. Panelbase consistently show the narrowest leads – for much of last year they were showing leads of about 8-10 points, that’s now narrowed to five points in their last two polls. Survation and ICM are also showing single point leads. YouGov and TNS are currently show leads in the mid-teens (though TNS has far more don’t knows than other companies), MORI continue to show leads up in the twenties. A broad range of leads, but the general trend across pollsters in recent months seems to have been a slow drift towards YES. All the polls so far are collected here.

On other matters, Populus’s twice weekly poll this morning had topline voting intentions of CON 34%, LAB 37%, LD 9%, UKIP 14%. Tabs here.


I’ve been entrenched in preparing Nick v Nigel debate polling (more on that later) for the last couple of day, but in the meantime there have been two new Scottish referendum polls.

The new YouGov poll in the Times this morning continues the recent trend of movement towards the YES camp. Topline figures are YES 37%(+2), NO 52%(-1), equating to to 42/58 split once don’t knows and won’t says are removed. YouGov tend to show one of the bigger NO leads, but the trend is there. Looking at the longer term figures from each company, back in September YouGov had YES at 38%, 39% in December and January, 40% in February, now 42% in March. YES have got a lot of catching up, progress is slow and there’s not that long to go, but the movement is there. Tabs for the new YouGov poll are here.

The second poll from TNS BMRB is contrasting, but is actually rather less new – fieldwork finished on the 9th March (I assume the long gap between fieldwork and publication for TNS polls is something to do with their face-to-face fieldwork, but it still seems to take a long time. Even face-to-face fieldwork these days is done on a laptop, so it’s not like lots of data needs to be collated by hand). Topline figures there are YES 28%(-1), NO 42%(nc). No continuation of the trend towards yes there.


There is a new Panelbase Scottish poll for NewsNet Scotland out today, with topline figures of YES 40%, NO 45%, don’t know 15%. Leaving aside that one poll with leading questions, the five point lead for No is the smallest we’ve seen since way back in February 2012. As ever don’t get too excited about any single poll, it’s the trend that counts, but there does appear to be a slight trend towards YES. John Curtice has his say upon it here.

Polls so far are here. Different Scottish polls from different companies tend to produce slightly different figures, especially in terms of don’t know. For trends it’s probably best to repercentage to exclude don’t knows, and one should certainly only compare polls from the same company:

Taking them one at a time, ICM had YES on 40% last September, then 46% in January, then 43% in February – YES appear up on September, but recent trends are unclear.

Ipsos MORI we had YES on 34% last September, 37% in December, 36% in February. Again YES appear up on September, but the recent trends are unclear

Survation we had YES on 38% in January, and then on 45% in February and March… but there was a significant methodology change between January and February, so don’t read too much into that shift.

TNS-BMRB we have what looks like a trend. YES was on 36% in October, 38% in November, 40% in December and January, 41% in February.

YouGov appears to show a similar steady but slow trend – 38% in September, 39% in December and January, 40% in February.

Panelbase have consistently shown better scores for YES than other companies, but until today have not really shown a clear trend: 44% in September, 45% in October and November, 43% and 44% in February. Repercentaged to exclude don’t knows today’s YES figure would be 47%… so higher, but not something that couldn’t be normal margin of error.

Putting it all together whatever trend is present is only small, so in individual poll series it is difficult to distinguish it from normal sample variation. Looking across the board though, the direction of travel in recent months does appear to slightly be towards YES.


Saturday’s Sun newspaper had a new YouGov poll of Scottish voting intentions with topline figures of YES 35%(+1), NO 53%(+1). There’s obviously no significant change from YouGov’s previous poll, carried out at the start of February before the recent currency row, and the NO lead remains at 18 points. The referendum results are here and there are some additional questions here. In his commentary on the YouGov website today Peter Kellner suggests views are pretty solid – the currency story hasn’t made much impact because the vast majority (79%) of YES supporters just didn’t believe it and assumed the British political parties were bluffing (though a fair amount of YES supporters would also prefer an independent Scottish currency anyway).

There was a similar break when people were asked about an independent Scotland’s position in the European Union – the large majority (70%) of YES supporters think that an independent Scotland will be able to make a smooth transition to membership on day one, only 15% of NO voters think they would. The arguments that dominate the Scottish independence debate don’t really appear to be changing any minds, people are just viewing them through their pre-existing support for YES or NO.

In a similar vein there is a new Ipsos MORI Scottish poll, also timed to mark the 200 days to go point, and again showing very little change. Amongst those certain to vote YES is on 32%(-2), NO is on 57%(nc), Don’t knows 11%. Changes are from the previous MORI poll in December 2013. Full tabs are here.