People lie to pollsters – regularly and consistently. How do we know that?
I’ll give you one example.
The Seanad abolition referendum in 2013 was preceded by opinion polls in which the great majority (more than 80%) of those polled told the pollsters that they would definitely be voting. And that they would be voting overwhelmingly in favour of abolition.
In the end just 39.8% of voters turned out.
But worse than that, within a fortnight after the referendum proposal was defeated by 51% to 49%, the Referendum Commission carried out a “scientific” poll to analyse the result. And the survey showed that 52% of people were now claiming that they had cast a vote.
So a good proportion of those surveyed were probably lying about the fact that they hadn’t voted at all.
Stranger still, the 52% of those surveyed who now claimed to have voted told the pollsters that 59% of them had voted “No”, while only 37% of them now claimed to have voted “Yes”. ( 4% refused to say how they had voted ).
If the people surveyed were remotely truthful, the referendum would have been defeated in a landslide.
Not merely did the pre-referendum polls wrongly suggest that the abolition of the Seanad would be carried by a two-to-one landslide; the post referendum polls wrongly indicated that it had been defeated by a 60-40 landslide. Exit polls conducted sporadically here and there were much more accurate.
Lying to pollsters – it’s not unusual
So people lie to pollsters as to whether they will turn out to vote and people lie to pollsters about whether they have voted and how they voted.
Is this unusual? No. Right across the world, pollsters are deceived. In 1961, a great many more people told pollsters that they had voted for John F. Kennedy than actually had. The same applied more recently to Barack Obama (above with David Cameron). The secret intentions of the British electorate confounded the pollsters in the last Westminster election.
There is a noticeable worldwide tendency to falsely claim in retrospect to have voted for the winning sides in elections and referendums.
Another issue in opinion polls is the fact that the pollster “comes to the person polled” rather than the voter getting up off his or her posterior and going down to the polling station to vote.
In order to correctly predict the outcome of an election by reference to polls, you always have to “factor in” the determination of different cohorts of people to go out and cast a vote. The Seanad referendum outcome showed the “No” voters polled were much more determined to vote than the “Yes” voters polled.
As I wrote in this blog recently, there will in truth be 40 general elections on February 26. And a crucial question will be “Who turns up to vote?” in each of those constituencies.
Will the angry vote and the protest vote turn out – especially if they twig that Enda is going to be Taoiseach no matter how they vote? Or will that make them more determined to kick over the traces? Or will the “steady as she goes” vote turn out? Will the young vote turn out to the same extent as they did in the Marriage Equality referendum? Or will it be the older voters who as usual turn out?
These are the crucial questions which lie at the heart of the three-week campaign.
What would Connolly and Larkin think?
The IPSOS/MRBI poll in the Irish Times published on the first day of the actual election campaign makes very interesting reading. If that poll translates into the result on polling day, the FG-Labour coalition is going to be defeated.
Unless there is a very strong swing towards the coalition parties in terms of turn-out on polling day they will struggle to get 70 or 72 seats, well short of the 79 seats needed for a Dáil majority.
It is “early days yet” in respect of this election race. But Fianna Fáil are already showing 21% support. The “toxic” label is not sticking to them. If older voters turn out more heavily, FF may well edge towards the 25% result they got in the local elections. That would give them 34 to 38 seats. Are there FF voters, like Tory voters in Britain, who are keeping mum to the pollsters?
Likewise the picture on the “Left” is intriguing. The Irish Times poll shows Labour getting 11% of AB voters. But given a national support level of 7%, this suggests that Labour could face wipe-out in traditional working-class areas. Fancy that for the party of Connolly and Larkin! Their support base has migrated to the comparatively well off middle-class liberal section of the electorate.
Sinn Féin faces two issues. Will their supporters turn out as voters? Will their candidates attract second preferences to pick up seats where they do not reach near the quota on the first count? But the Irish Times poll would suggest that they will trounce Labour in terms of working class support.
One more warning. National opinion polls seek out support only by party label. Voters may, however, vote by reference to the high profile of candidates.
Another thing to bear in mind is the trend in opinion polls. That may be more significant than the absolute levels of support shown in any one poll.
The Irish Times poll was conducted in the context of an imminent election. It showed no trend at all in opinion towards the re-election of the current coalition. The “vote for us to keep the recovery going” message has not yet shown any traction.
Unless a strong trend in favour of FG and Labour emerges during the campaign, the FG-FF coalition seems the most likely outcome.
Both FF and FG will furiously disown such a result – until it happens! And then the “national interest” kicks into play.
My advice to punters on this election race is to remember that there are 40 separate races. Examine the runners in each race very carefully. Go by form. Don’t go by the pollster tipsters. Consider the going. Do the sums carefully. And trust your own judgment.
Michael McDowell is a senior counsel (and a very good one, at that) and former TD who served as Justice Minister and Tanaiste, as well as leader of the PDs. He will contribute to the Paddy Power Blog during the 2016 Irish General Election.
- Michael McDowell: Enda Kenny WILL be Taoiseach again but based on Labour’s collapse here’s how the next Government could shape up
- Michael McDowell: Micheál Martin’s leadership of Fianna Fail and personal career could end in a matter of weeks
- Ivan Yates: Here’s how Labour could lose 25 seats in the Irish General Election
- Ivan Yates: How resurgent Haughey and Hanafin can get Fianna Fail up to 40 seats in the Irish General Election
- Ivan Yates: Sinn Fein are no longer toxic – here’s how they can take up to 30 seats in the Irish General Election
- Liz O’Donnell: There’s no place in Government for an inherently awkward party prone to hissy fits