Survation Poll for ITV/Daily Mirror puts the Kippers just 1% behind the Tories

[-:Undertaker:-]

http://survation.com/2013/07/latest-...emains-strong/
http://order-order.com/2013/07/05/uk...urvation-poll/

Latest Voting Intention Results – Tories fall to record low in polling results as support for UKIP remains strong.
July 5th, 2013

Survation poll for The Daily Mirror / ITV Daybreak on Friday 5 July. Topic: Public attitudes towards the NHS. Fieldwork: 3rd July. 1,085 UK respondents.

Headline Voting Intention Details (changes since June 2nd)

Conservative: 23% (-2%)

Labour: 36% (nc)

Liberal Democrats: 10% (nc)

UKIP: 22% (+2%)

Others: 10% (nc)

Analysis and comparative methodology

Despite recent talk of a possible improvement in Conservative Party fortunes, the latest Survation poll continues to show the party at record lows of support. At 23% this is the lowest Conservative voting intention figure published by a BPC member so far this Parliament. The apparent reason, evident from the figures, is the continued record-high polling for the UK Independence Party (UKIP), here scoring at 22%. This is within margin of error of the Conservative Party, suggesting it is possible that UKIP is actually ahead of the Conservatives (though conversely it is possible that the gap between them is a few points higher than the 1% shown).

Voters Need A Reminder Of Who Might Be On Their Ballot Paper

Since their high-profile performance at the local elections back in May, which gave the party a major publicity boost, UKIP has been polling steadily around the 20% mark in Survation polls. Whilst some other polling companies have shown a more recent fall-off in UKIP vote, this is likely a result of methodological differences, where failing to prompt (initially mention the party in the voting question) for UKIP causes people to forget about them as a valid option when considering a voting intention question.

News coverage of the party has abated somewhat over the last month as media interest from the local elections has faded. This would account for a “fall in support” when the party are not prompted for, compared with Survation polls (which include UKIP in the voting question prompt) that have shown their support holding up.

Simply put, not mentioning UKIP in the voting question lowers their apparent support, particularly in periods such as this, when the party has not featured as heavily in the news media.

Why is YouGov The Online Outlier?

YouGov in particular publishes significantly lower UKIP figures in it’s daily polls (today’s 12% making it an outlier among online pollsters), largely as a result of methodological differences such as its use of Party ID weighting to a 3 year old target, which after all usual weighting has taken place, reduces UKIP’s figure significantly. Other online companies such as Opinium and TNS show UKIP at 19% in their most recent polls, more comparable with the Survation figures, though still with a slight difference perhaps accounted for by prompting (as discussed) or by other minor differences in weighting.

YouGov’s President, Peter Kellner has perfomed an analysis of the effect of prompting, showing that if YouGov used Survation’s methodology, their results would look similar in a detailed piece for those interested here:
- See more at: http://survation.com/2013/07/latest-....iSxODpDp.dpuf

Interesting the gap that has opened up between some of the polling companies. I tend to side with Survation etc who prompt UKIP (as I naturally want them to be doing well) but also from a polling perspective - as the polling expert Mike Smithson of political betting has commented before, YouGov applying the same weights from 3 years ago in a three-party system to todays four-party system can yield very wrong results indeed (see Eastleigh and the Local Council Elections in 2013).

But yes, interesting. We'll only know for sure come May 2015.

Thoughts?

[Chippiewill]

Poor methodology reveals outlandish polling results? Shock horror!

[FlyingJesus]

Polls that prompt for a specific answer get that answer, huge news. Not really sure that asking 1000 people "do you think UKIP would be good for the NHS" yields a likely general election outcome for the entire country

[-:Undertaker:-]

Poor methodology reveals outlandish polling results? Shock horror!

You've stated in the past that you disagree with past vote weighting (which YouGov does) so I assume you are talking of YouGov as the 'outlandish result'?

Polls that prompt for a specific answer get that answer, huge news. Not really sure that asking 1000 people "do you think UKIP would be good for the NHS" yields a likely general election outcome for the entire country

You have grasped the wrong end of the stick.

The difference isn't in a worded answer (as there isn't one) - the difference is simply whether to list UKIP alongside Con, Lab and LDEM when asked 'Who do you intend to vote for in the General Election come 2015?'

That's what is causing the huge difference in data, which you'd understand had you read properly.

[Chippiewill]

You've stated in the past that you disagree with past vote weighting (which YouGov does) so I assume you are talking of YouGov as the 'outlandish result'?

Wasn't saying YouGov was perfect either, it's certainly better than outright ignoring demographics.

[-:Undertaker:-]

Wasn't saying YouGov was perfect either, it's certainly better than outright ignoring demographics.

As far as i'm aware, Survation doesn't ignore demographics (unless you can find source otherwise saying so?) and the issue or divide in the polling between the companies is mainly based in reallocating the 'don't knows' based on 2010 results which was then a three-party system.

..as always, best way to check is against real electoral results. Here are the one's from Friday 27th June 2013 -

http://ukgeneralelection2015.blogspo....html?spref=tw

UKIP performances in yesterday's by-elections:

BASILDON - Billericay East: 29.84%

DARTFORD - Newtown: 22.71%

PLYMOUTH - Southway: 26.45%

RUTLAND - Ketton: 17.47%

SOUTH TYNESIDE - Cleadon and East Boldon: 26.06%

SOUTH TYNESIDE - Primrose: 34.64%

WORCESTERSHIRE - Stourport-on-Severn: 25.97%

Click the link below if you want to see all the results in full
8 by election results 27th June

UKIP overall performance in contested by-elections yesterday:

UKIP votes: 3,705
Total votes: 13,863
UKIP percentage: 26.73%

UKIP overall performance, including Newark & Sherwood where they didn't stand:

UKIP votes: 3,705
Total votes: 15,719
UKIP percentage: 23.57%

[Chippiewill]

As far as i'm aware, Survation doesn't ignore demographics

Not forcibly so, but surveying at certain times, certain locations will result in a demographic bias.

[-:Undertaker:-]

Not forcibly so, but surveying at certain times, certain locations will result in a demographic bias.

Which is why it's a member of the British Polling Council (BPC) because it's a respected pollster and not your local newspaper online poll.

[Chippiewill]

UK Exam Boards are part of JCQ, doesn't mean that they're necessarily any good. Likewise with polling firms.

[-:Undertaker:-]

UK Exam Boards are part of JCQ, doesn't mean that they're necessarily any good. Likewise with polling firms.

True, and that's the debate - are Survation more accurate when it comes to this or YouGov? I would used to have said YouGov, but as explained - it all seems to have changed now and Survation's polls are being backed up by real results every week in the mid-20s upwards for the purples.