Beware the Disconnected and Undecided Voter
On the face it, the last couple of weeks have been positive for the Repeal side of the upcoming 8th Amendment Referendum. First the Oireachtas committee recommendations have backed up the findings of the Citizen Assembly and recommended repealing the 8th Amendment and suggested that legislation should allow for abortion on demand for pregnancies up to 12 weeks gestation. Following this several TD’s having studied the evidence, have backed the Oireachtas stand; not the least of whom was Michael Martin in his address to the Dail during the debate last week, a significant turnaround from his previous stance.
This has been followed by two opinions polls that appear to show a clear lead for both the repeal of the 8th Amendment and the introduction of legislation for abortion to be allowed on demand up to 12 weeks. In our poll this weekend we split the questions to firstly determine the level of support for repealing the 8th Amendment, and then separately asked about support for this to be replaced by legislation for abortion on demand up to 12 weeks.
On the appeal of the 8th Amendment 60% support Repeal, with 20% not supporting repeal, while 20% remain undecided. With undecided voters removed this means suggest a 75/25 ratio of Repeal to Retain voters in the population as whole.
When this is taken to the next step however, and voters are asked how they would vote in terms of potentially replacing the 8th Amendment with legislation to allow abortion on demand in the first 12 weeks, 51% support, 27% do not support and 22% remain undecided. Again, removing the undecided voters would suggest 66/34 ratio in favour of allowing abortion on demand up to 12 weeks. This reduced level of support is due to the fact that 1 in 5 of those who agree we need to repeal the 8th Amendment, do NOT agree to abortion on demand.
If you are to simply remove the undecided voters as we have done above, this does appear to show a very strong lead for both repeal and replacement with legislation for abortion on demand in the first 12 weeks. The problem with this assumption is threefold a) voter understanding, b) how undecided voters will distribute and c) which voters are more likely to turnout.
Firstly, there is voter understanding. The whole concept of the 8th Amendment being in the constitution vs. a replacement that isn’t, is quite complicated for a lot of voters who are perhaps not really that close to and don’t yet understand all the nuances and legal issues behind this. The members of Citizens Assembly, the Oireachtas Committee, and many TD’s have studied all sides of the arguments and been explained all the issues in a deliberative context. The same cannot be said for all the members of the general public. Some of course, are very interested in the topic and will have read all the reports on the issue, but there are also many people who will simply not have engaged with the issue yet. In most of the work we have done post referendums, the general public often do not really engage in the issues until the start of the 3-4 weeks campaign before the referendum itself.
Secondly, the strong figures for the Repeal side, assume that all undecided voters are the same as those that are decided, and that their votes will fall in the same proportion when they do make up their mind. A significant body of analyses conducted by RED C, over a series of referendums, suggests this often is not the case.
When voters are asked to change the constitution, but are not sure about whether change is a good idea, on the whole they tend to revert to the status quo. Simply expressed, this behaviour lies in the fact that people often decide it is better not to change something, if they themselves are not impacted by what is currently in place, and are also not sure if the alternative will be any better.
There is also the “shy” voter factor. This looks at the fact that when it is clear that the popular view is heading one way in a referendum, those voters with the opposing view become somewhat “shy” in telling people how they really feel. Most of the recent coverage on the 8th Amendment suggests a move toward the Repeal side, so if you don’t agree with that position you may become somewhat shy about saying so.
In the Marriage Equality referendum, RED C saw exactly this effect come into play, and ended up using a wisdom of crowds approach to overcome this, and correctly predict the result. Wisdom of Crowds overcomes the shy effect as it asks people how they think “others” will vote, thus removing the shy effect. In the Marriage Equality referendum when we compared claimed personal behaviour, to the prediction of others, it was clear that the majority of those claiming they were undecided in the final stages of the campaign actually ended up voting No.
Finally, there is the issue of turnout. These figures assume that everyone is as likely to turn out and vote for the referendum. But in reality, those whose views are more strongly felt are much more likely to turn out, than those who are the middle ground. There is evidence to suggest that those opposed to repeal, may be more likely to turn out in total, while the larger group currently support repeal may not be as motivated.
What does that mean for todays poll results? Well firstly, they should be taken with some caution, as we are still some way out from the referendum itself, and voters may not yet be connected enough to give an opinion based on a real understanding of the issues and how they will vote. Secondly, historical precedence and past referendum analysis suggests, that if anything, a greater proportion of those claiming they are undecided at this stage are likely to vote No rather than Yes. Thirdly, those on the No side are perhaps more likely to turn out and vote.
That means the gap between the two sides is likely to be much closer than the raw figures suggest. In fact, if you were to use the Marriage Referendum as an example, you could almost re-allocate all undecided voters to the No side and then you would be looking at very small margins between the two sides. With 60:40 for suggesting a straight repeal vote might still pass, but with 51:49 for legislation on demand in the first 12 weeks suggesting that currently if this is proposed as part of any referendum, it may not.
Download Report Below: