Brexit vote: Should it have come as such a shock?
Belfast pollster Bill White examines whether the UK's decision to leave the European Union in a referendum early this year should have been such a shock.
We're still getting the constant refrain from the media that the UK EU Referendum Brexit result was a surprise, it was a bolt-from-the-blue, and no-one saw it coming. Really?
We even now have Mr Trump himself saying there's going to be a 'Brexit' in the US election ie he's going to win and it's going to be a big surprise. Why is this? - Were the polls that wrong in the EU Referendum?
As a pollster we constantly track all the polls across the UK and the US and indeed usually put up updates on LucidTalk's social media accounts. So I decided to review what LucidTalk's Twitter and Facebook were reporting about the polls in the weeks before the EU Referendum on June 23rd.
A table of all the UK-wide EU Referendum polls between Mid-May to Mid-June was published on our social media accounts on the 14th June i.e. 10 days before the actual referendum - see below.
As you will see this table also shows whether the polls were carried out online or by phone.
You will see in particular the accuracy of the online polls! At this stage, between 6 weeks to 2 weeks out from the referendum the online polls showed a 2% to 5% Leave lead - the final result was 52% Leave to 48% Remain i.e. a 4% point Leave lead. Even here in little old Norn Ireland LucidTalk's online polling predicted the final NI result to within 1% in the NI SUN newspaper on the Monday (20th June) of the referendum week. In fact, if you average all the online polling in the above table, across this 4 week period (mid May to mid June) it comes in at 51% Leave to 49% Remain - not a bad forecast in terms of the actual final result!
Even leaving aside the undoubted better accuracy of the online polls, the average of all the polls (telephone and online) were showing a tight race with the forecast being within 53/47 to 47/53 either way. So basically taking into account the normal polling margin of error, the result was always going to be a toss-up, i.e. too close to call!
So with the accuracy of all the polling, I asked myself why the result was such a surprise to a lot of people? - frankly including myself.
I came to the conclusion that it was the way the media reported and spun the results.
The media had an undoubted pro Remain bias which ironically the post EU referendum polling showed had a sort of reverse effect, and contributed to the swing from Remain to Leave in the last months of the campaign.
Likewise the Remain campaign was so poor, both here and in Great Britain that it also had this sort of reverse affect - As the Irish Times Pat Leahy put it 'Every time the Remain campaign said we have a panel of experts, another Leave voter was born'!
The key thing with polls is to view them in the round, taking into account all the other evidence and factors that are available at the time.
Merge the poll findings in with your own experience, read the poll reviews and commentaries, ask other people about the poll results - what do they think?, and do the poll results feel right to you?
Above all read the actual poll results, and not just what the media headlines say about the poll results. Look at what the table above says (published by LucidTalk 10 days before the referendum) and ask yourself 'Why the surprise'?
*Bill White is Managing Director of Belfast polling and market research company LucidTalk. You can follow LucidTalk on Twitter: @LucidTalk.