Do opinion polls influence the outcome of elections?
Last year pollsters failed spectacularly to get anywhere close to predicting the General Election's final outcome
The general election 2015 wasn't exactly a success for the UK polling industry, and although LucidTalk had no involvement with the national UK polling, we still get the little jibe remarks about our own polls, and 'getting it wrong', but we take all that in good spirit.
After last year's general election, the British Polling Council initiated an independent inquiry which published their interim findings a couple of weeks back - you may have seen reports about this in the media.
Basically they found that Conservative voters were harder to detect than Labour voters - now there's a surprise - and polling therefore needs to be carried out over a longer period, to track opinion accurately and detect these 'shy Tory voters'.
The full report and recommendations from the inquiry will be published in March and are expected to recommend the concept of 'super-polls' involving perhaps higher sample sizes, and longer polling periods of say five to six days.
Instead of the media pressurised one to two day polls that are currently the norm.
A longer poll period should help with reaching crucial 'hard to reach' voting groups more effectively, such as e.g. the over 70s.
Maybe more by telephone, as this group are less likely to be internet connected (for online polling).
However there's an additional question - did the polling affect the outcome of the election?
There are a lot of Labour and LibDem people who think the polls certainly did have an effect. Remember the line - 'Hung parliament with Labour minority, and Nicola Sturgeon will run the country'.
This line certainly has more potency when the polls say the SNP could've held the balance in a hung parliament situation.
Had the country known that the Conservatives were in the lead, maybe that would have placed the Conservatives under more scrutiny.
Mind you, all of this is 'matter of opinion' - there's no way of proving this one way or the other, unless we could re-run the election again with polls more reflective of what the final result turned out to be.
Now of course, the general election polling has lead to the politicians trying to interfere again (as they usually do when they see something they can get media coverage for).
The Scots Labour peer Lord Foulkes is introducing a private member's bill in the Lords to try and regulate UK polling.
Of course this is all total nonsense (as Lord Foulkes knows it is) as in a free society you can't regulate the collection of opinions plus in any case, well-funded privileged minorities such as banks could still conduct private polling and influence the flow of an election campaign that way.
Plus, if anything along these lines did become law, there would be no way of stopping someone (like me) setting up office in say Monaghan and providing a service to all the big UK poll companies by publishing their results from there.
The more important point is how we all should treat polls and polling.
I said in a previous article that everyone is entitled to take out, or reject, whatever they want from poll results. I work in the industry and there's plenty of polling that I reject because it doesn't fit-in to what I feel is right.
Frankly, a good example is our own LucidTalk NI polling regarding the NI 'same-sex marriage' issue which showed 37% of DUP supporters would support same-sex marriage (this also fits-in with polls carried out by other companies).
Do I believe this? - No.
I do believe a majority of these DUP poll respondents currently honestly 'think' they would support same-sex marriage if it came to a referendum on the issue - but when hardy-came-to-hardy I don't think they would.
However, I say again, this is just my opinion, and my interpretation of our own poll results - I could be wrong.
The main key advantage of polling is that it allows us to get regular snap-shots of public opinion, about not only voting intention, but views on policies, political leaders, and a whole host of topics that can't be measured in public elections.
They also allow us to see how males/females, particular age-groups, and here in NI how our religious groups think about topics and issues - again these can't be measured just by public elections.
Also how are we to get reliable information as to what the public are thinking about? If not polling, what? Twitter and Facebook? Or a handful of journalists who all get their information from each other and their own clique of political contacts?
Plus locally (or nationally - Ireland or UK) we can't depend on politicians to truthfully tell us how an election campaign is going, or what issues are important, either.
Watch as the Northern Ireland Assembly election campaign unfolds and the political parties predictions as to how well they are going to do.
If you add all their predictions together we won't have 108 MLAs elected but probably somewhere around 210!
And watch out for 1,000s of renditions on social media from our aspiring politicians of that line 'GReeaaat reception on the doorsteps tonight!'.
Finally, and before you (and us) continue to knock the polls too much - we can still say that pollsters are never wrong if you follow the two 'Golden-Rule Poll questions':
- Which political party is best for the economy?
- And which political leader would make the best Prime Minister?
No party or leader has ever won an election being behind on both these questions in the polls, and the 2015 general election was no exception - Labour were always behind the Conservatives on the economy, and Miliband was always behind Cameron as the best PM.
So the signs were all there, and the polls were in fact right - if only we, the media, and also incidentally the polling companies as well, had followed the correct poll results.
Bill White is Managing Director of Belfast polling and market research company LucidTalk. You can follow LucidTalk on Twitter: @LucidTalk.