Marriage referendum: Yes lead narrows but still enough to win

Slow drift in Yes vote but public mindset seems less volatile than in other votes

The latest IrishTimes/Ipsos MRBI same-sex marriage referendum poll predictably shows a softening of the Yes vote: excluding undecideds, 70 per cent are in favour of a change, with 30 per cent against.

The Yes margin is comfortable and most of the indicators point to a firming up of the Yes vote as opposed to a falling apart. That said, recent referendums have shown no lead is unassailable.

The final margin in favour of a Yes vote will be narrower if the trend continues and the pattern of previous referendums is repeated.

Yes anxiety

Referendums with a social or moral dimension rarely pull quietly into the station. The 1995 divorce referendum gave us our first taste of Yes voter anxiety. A lead of 44 points for the Yes side all but evaporated as the campaign played out. In the end, divorce made it across the line, just.


The 2012 children’s referendum result confounded all polls with the speed and severity of the shift in preferences. Similarly, the referendum to abolish the Seanad, ultimately defeated, was all dive, no drift.

All the signs from Saturday’s poll point towards a less volatile voter mindset. The level of undecided at this stage is relatively low (at 13 per cent). While some voters feel they are unable to openly discuss their views, the issues have been widely debated in the media. There is no empty space to be colonised by new arguments.

And voters claim to have largely made up their minds. Among No voters, only one in five (22 per cent) is still listening to both sides of the debate. Yes voters (just 19 per cent still listening) are even more assured.

So for this same-sex marriage referendum, the data is pointing to drift, not dive.

But there other threats to a Yes vote revealed in the latest poll findings. Turnout is always influential in the result. Typically, certain demographic groups such as older or more affluent voters are more likely to vote in referendums.

How differential turnout will affect the outcome is not obvious. More older voters turning out will likely narrow the margin (among those aged 65 and older, 60 per cent intend to vote No), whereas a higher turnout in more affluent constituencies will widen the margin (ABs are 82 per cent Yes).

To further complicate matters, younger voters are more likely to be affected themselves or have a friend or family member affected by the outcome (42 per cent for under-35s) than older voters (25 per cent for 35+ year-olds) and this could be the incentive some younger voters need to get out and vote.

How the undecideds fall has the potential to narrow the gap, although with relatively few undecideds they may play a lesser role than in previous referendums.

The conventional wisdom is undecideds are more likely to favour the status quo. In this latest poll we measured attitudes towards same-sex marriage in addition to voting intentions. Our analysis reveals a pattern of responses that is consistent with the view that undecideds lean towards No.

No mindset

More undecideds display a No mindset than a Yes mindset. For example, more undecideds believe same-sex marriage will diminish the institution of marriage than think it will enhance it. On this analysis, it would be reasonable to expect more undecideds to come down on the side of a No vote.

Conventional wisdom is also that voters can use a referendum to give the Government a bloody nose. While there is no way to isolate this motivation within the data, the poll shows a very negative reaction to the referendum on the eligible age for a president: just 30 per cent are in favour.