As with every other Scottish subsample to have been published by YouGov since the referendum, today's has the SNP in the lead. However, the gap is towards the lower end of the normal range we've become used to...
SNP 34%, Labour 28%, Conservatives 16%, UKIP 11%, Liberal Democrats 9%, Greens 2%.
Doubtless one or two of the minions at McDougall Central will be hoping that this is an early sign that Jackanory Jim's bizarre antics might finally be gaining some traction with the public. I think that's unlikely, though, because Labour's vote share isn't untypically high. Instead there are more UKIP supporters than usual in the sample, plus the Liberal Democrats are doing a little better.
We can probably expect bigger fluctuations from day to day now that YouGov are relying on much smaller Scottish samples - for the first time since heaven-knows-when, the whole Scottish sample has actually had to be upweighted in today's poll (from 139 to 149). In spite of that, SNP identifiers have still been downweighted, albeit by less than usual.
We've now had enough polls this week to make it just about worthwhile updating the Poll of Polls for the first time in 2015. Take the following numbers with an extremely heavy dose of salt, though, because they're based on just four subsamples - three from YouGov and one from Populus. As always, the Opinium poll can't be used because the Scottish subsample wasn't published.
Scottish voting intentions for the May 2015 UK general election :
SNP 39.8% (-4.3)
Labour 27.3% (+0.8)
Conservatives 15.5% (-1.0)
Liberal Democrats 6.8% (+1.0)
Greens 5.5% (+3.8)
UKIP 4.5% (+0.5)
(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.)
Does look like Yougov are trying to reduce respondent numbers in Scotland by sending out less invites and, presumably, changing the targeting (in what way we might ask?).
ReplyDeleteIt’s an odd thing to do. For a pollster, having too many respondents from a group (e.g. Scots) is great. It’s exactly what you want. Having too few is a huge problem as you start having to invent their views based on those of the small sample you have managed to get.
If say you were getting flooded with SNP respondents that’s not a worry; it’s what weighting is for. You just get a great picture of the mind of SNP respondents. So long as you are getting a good sample of other voters, extra from enthusiastic voters of one party just isn't an issue. It’s like the older vs younger or ABC1 vs C2DE thing. Pollsters normally have lots of respondents from the first group in each case, but fret about their findings for the second groups as they can struggle to reach them.
Yougov haven’t been getting flooded by SNP anyway, as Labour were well ahead in their subsets prior to the iref, yet the number of Scottish respondents was no different to post-iref; consistently a little higher than the rUK per capita and what we can assume was increased engagement on all sides. That and historically, in contrast to other pollsters, Yougov seemed to have Labour over-represented in their panels with a lack of SNP respondents based on Scotland-only polls.
Anyway, if up-weighting of the Scottish sample numbers becomes a common feature, we’ll have to start considering yougov UK poll Scotland subsets as questionable due to the invention of VI it will inevitably be creating. Certainly, if anything, such up-weighting will increase volatility, as is common to any group where you struggle to get respondents.
I'm I right in thinking that the SNP have gone from 38 to 34% and labour from 23% to 28%? In the space of 24hrs from the same pollster?
ReplyDeleteAnd they want us to take them seriously?
There's nothing unusual or suspicious about that - these are subsamples, not full-scale polls.
DeleteI think it's interesting that Ofcom have today ruled that UKIP are a "major party" in England and Wales, but not in Scotland or NI. I wonder if this will cause the broadcaster a problem in respect of showing TV debates (if there are any) involving UKIP in Scotland.
ReplyDeleteThe logic of excluding a "major" party from UK debates on the grounds that they are only major in Scotland, Wales or NI is just about defendable, but showing a debate involving one party that is ruled not to be "major" is on even dodgier ground.
Interesting comment from Anthony Wells at Yougov:
ReplyDeletehttp://ukpollingreport.co.uk/blog/archives/9127/comment-page-5#comment-959659
"Couper – fairly simple actually. We used to sample all of GB together, and just rely on Scotland falling out in roughly the right proportions naturally. Having the SNP as a target weight obviously helped iron out some variation, but it was still very much in the lap of the gods and was often a little too Conservative. The change is to sample most Scottish respondents separately, using similar quotas as those we do for proper bespoke Scottish polls.
Yesterday’s was probably just a weird sample – you get these things, especially with a sample of 139, which is why I’m always begging people to ignore single crossbreaks!"
The weird sample is the one above with the Lib dem vote shooting up and Farage riding a Scottish wave.
So in that sense what they've done should probably be ok. Time will tell if it really changes anything much. Although in the end, it is full Scottish polls that matter.
I hope to the gods that UKIP have been seriously overestimated in these subsamples...
ReplyDeleteI agree with the sentiment,I hope many who said they would vote for them were just a bit drunk at the time,and wont so making it even more of a freak poll.
Delete39-27 in the crossbreak poll of polls.
ReplyDeleteSqueaky bum time for SNP supporters?
On 46% today; double that of Labour on 24%, so apparently not.
Delete38-27 in today's Populus poll.
DeleteWhat happens to the "don't knows" and "wont says" in these polls? Do they just get ignored? Genuinely curious.
Yougov and populus can't be compared. Apples and coconuts.
DeleteAlso 38% after being down-weighted by up to 39% (x0.39) using 2010 weighting; something even Prof C. says is disastrously inaccurate and results in underestimation of SNP VI.
There's always some don't knows. The bulk of them don't / won't vote. Good way of knowing this lost is to ask them. If people tell you they didn't vote in the last election, you can be very sure it's the truth.
Some DKs who have voted before and say they are likely to vote are usually reallocated in part to what they voted last time. That would most likely benefit Labour who got the most votes last time.
Thanks for the that substantial and informative reply. I find all this polling stuff quite fascinating! Do you have any idea when the Ashcroft polling for Scotland's constituencies will be published? This will surely give a much more accurate picture of what we can expect from the election.
Delete