Archive for the 'Pollsters/polling' Category

h1

STAY likely to win the EU In/Out referendum for the same reason that CON won GE15 – the fear of the unknown

Saturday, May 23rd, 2015

On the face of it the numbers look good for STAY but are they?

One of the things that the Tory victory on May 7th ensures is that during this parliament there will be an in/out referendum on Britain’s membership of the EU. The question is which way will it go?

A big campaigning lesson from the general election was how successfully the Tories were able to deploy the fear of the unknown factor in the closing few days. Labour and the Lib Dems in seats they were defending were simply not prepared and had no answer.

The same, I’d suggest, could happen in the coming EU referendum. When faced with a choice between the status quo and the unknown British voters have a long record of opting for the former. Fears about what tomorrow could bring are a very powerful campaign message and will be used extensively by those wanting us to stay.

The polls above, only one of which was carried out after GE15, presents a fairly consistent picture although in today’s context people will rightly question the validity of all political polling.

A factor that could change everything, of course, will be how Cameron’s negotiations with other European leaders on the where Britain has specific concerns are seen to have gone. My guess is that he’ll seek to present the the outcome as showing that sufficient progress has been made to enable him to report things in a positive light. If Cameron recommends LEAVE then in the current context that could happen.

One problem with referenda is that the actual issue being voted upon is sidelined and it becomes a vote on something else.

The bookies make STAY the 4/7 favourite.

Mike Smithson





h1

Pollsters should follow Ipsos MORI’s 2008 example and not rush to resume general election surveys

Tuesday, May 19th, 2015

It had had a 4% Ken lead for mayor – Boris finished up 6% ahead

Back in 2008 at the first Boris-Ken battle the opinion polls became an issue during the campaign. YouGov was showing Boris leads close to what happened (a 6.2% Johnson lead) while Ipsos, in its final poll had Ken 3% ahead.

The polling led to a lot of attacks on YouGov from Ken and the Labour camp but in the end the online firm was vindicated.

Shortly after the results were in Ipsos-MORI announced that it would be reviewing its methodology and for the next couple of months or so while that was taking place it did not publish any voting polls.

The review led to two big changes. All voting intention polling was to be done by phone and a new weighting introduced for public sector workers who for some reason were being over-represented in its samples.

    In the current context a temporary period of few or no polls is, surely, the right way forward for the polling industry. Not one pollsterscame out of GE15 with its reputation undamaged.

My guess is that the industry needs a successful election, the EU referendum maybe, for confidence to be restored

YouGov appears to have suspended its daily polls for News International though it did publish a non-voting poll at the weekend for the Sunday Times covering a wide number of issues.

The Guardian has said that its long-running series with ICM will continue but findings will not be given too much prominence. Meanwhile we await the British Polling Council review which could take some time.

There’ll come a period in the near future when there’ll be a clamour for voting polls once again – maybe around the time of Mr. Osborne’s promised budget or when Labour’s new leader is in place.

Mike Smithson





h1

GE2015: The Inquest. A special podcast in collaboration with Polling Matters

Saturday, May 16th, 2015

IMG_20150513_104124 (1)

One of the great developments of the campaign has been the appearance of many new online resources for analysis and discussion. One of these has been the excellent Polling Matters created by Keiran Pedley of pollster GFK.

This afternoon, in collaboration with Polling Matters, we have a special podcast featuring some of the key players.

Keiran spent the last week speaking to several leading experts in the polling industry including Professor John Curtice, Lord Foulkes, Damian Lyons Lowe of Survation, Anthony Wells of YouGov, Matt Singh and Rob Vance. This podcast is a compilation of these interviews and is well worth a listen for those interested in what happened and where the polling industry goes from here in future.

Keiran tweets about politics and public opinion polling at @keiranpedley

Mike Smithson





h1

The real answer to the “shy Tories” phenomenon is Boris

Thursday, May 14th, 2015

In 2012 many more said they’d vote for him than actually did

A lot has been written in the past week about so-called “shy Tories” who are reluctant to tell pollsters on the phone or when they complete online questionnaires that they’ll support the blues.

Yet look at what happened in 2012 when Boris was re-elected as mayor of London – the opposite happened. Every single pollsters had Boris lead at a level bigger than it was.

Of the firms featuring in the chart only one used the phone at that election – Populus- and it came out worst of all.

Mike Smithson





h1

What went wrong with the polling? More starts to emerge but few answers

Tuesday, May 12th, 2015

In this News Statesman article James Morris, from Labour’s US-based pollster GQR, explain how its approach was different and is likely to produce far fewer don’t knows which, it is argued, add to accuracy. He also notes that I was one of those interviewed.

“The main difference between our polls and the newspaper polls is that we don’t ask the voting intention first. As Politicalbetting.com’s Mike Smithson found out when he accidentally participated in our only telephone poll of the last 4 years, we first ask respondents to think about the country, the economy, their top issues, the parties and the leaders. We think it gets them closer to their ballot box mindset.

This technique delivers a much lower don’t know number – generally half the level found in the public polls. We treat this ‘don’t know’ group differently to most of the public polls, asking them questions about who they are likely to vote for rather than assuming they are likely to vote for whoever they voted for last time. Of course, that requires many more questions and so is more expensive to implement especially for a phone pollster where every minute costs money. If we had run a final poll close to election day, would we have got the Tory margin right? It’s hard to know. But if this explanation is broadly true, it means the drift to online polling remains valid..”

Morris was one of the pollsters featuring in last night’s Newsnight examination of why the figures were so adrift. This is well worth watching.

Survation, which chickened out of publishing its final phone poll which had a 6% CON lead, is arguing that the polls weren’t wrong – there was just a very late swing. I think there might be something in that but none of the other pollsters found that movement in the closing hours last week.

It is hard to reach any conclusions on this. Getting representative samples is clearly an increasing challenge. Are those whom the phone pollsters manage to reach and are then ready to participate really representative of the electorate as a whole? As for the online firms does the fact that their numbers come from people who have volunteered to be part of a panel make them unrepresentative?

It was odd after the first debate that the online polls found that more in its samples were saying they watched than the official viewing figures suggested.

As we saw with the polling performances in Israel a few weeks back and at some recent US elections the challenge is not just a UK one.

Mike Smithson




h1

So were there really shy Kippers?

Sunday, May 10th, 2015

Prior to the election, there was speculation, the pollsters weren’t picking up “Shy Kippers”.

Though every single phone poll underestimated UKIP, it was within an acceptable amount, and the largest errors were from the online polls from Panelbase and Survation, who overestimated UKIP by 3%.

The polls that underestimated UKIP, on average, underestimated UKIP by 1.6%, and the polls that overestimated, did so on average by 1.9%. The online pollsters overestimated UKIP by 0.8% and the phone underestimated UKIP by 1.7%.

So for shy Kippers in the phone polls and overrepresented Kippers in the online polls, yes the data does confirm that, but given the smallness of the errors, people shouldn’t put too much faith in the Shy Kippers meme.

In what has been a difficult few days for the pollsters, they can take some comfort that they at least got the UKIP share of the vote largely right.

TSE



h1

CON moves to 9% lead in first GE2020 poll.

Saturday, May 9th, 2015



h1

Our GE15 prediction competition: results + chart showing CON leads % -/+ recorded in final polls

Saturday, May 9th, 2015

res
Full chart with all entries here

Sporting Index, which is sponsoring PB’s General Election coverage, is providing a competition prize of £200, payable into the winner’s SPIN account. If the winner does not have an account then he/she will have to open one to receive the prize. This is only open to people over the age of 18.

In the next few days the firm will have a bigger range of GE2015 markets up.

On top of that there will be a special prize – a signed Marf print of the winner’s choice of a GE15 cartoon that has appeared on the site.

Congratulations N Simms

Could you send me an email so I can arrange for the prize? Email.

Mike Smithson