ANALYSIS:Levels of 'knowledge' of the treaty suggest public opinion is more like Nice 1 than Nice 2, writes Richard Sinnott.
NOW THAT the Lisbon race is on, it is worth recalling a few basic facts.
In the first Nice referendum in June 2001, the No vote as a proportion of the electorate actually fell - from 21 per cent in the previous EU referendum to 19 per cent. But the Yes vote fell a lot more (from 34 per cent to 16 per cent) and so the No side won.
The main reason for the collapse of the Yes vote was the two-thirds of the electorate who stayed at home. Forty-four per cent of these non-voters gave lack of understanding/lack of knowledge as their reason for not voting. And the effects of lack of knowledge did not end there - 39 per cent who voted No in June 2001 cited lack of information as their reason.
So knowledge matters in EU referendums.
But knowledge in this context is as much or more a general state of mind as a grasp of a set of specific facts about the treaty - knowledge is measured on a subjective scale.
What matters is people's confidence that they know enough to make a decision. Because of this subjective element, the problem of low levels of knowledge is difficult to grapple with and the electorate's collective state of mind in this regard is probably not helped by all the talk (including in this article) about inadequate levels of knowledge.
That is not to say that the problem can be ignored. Arguably, it was lack of awareness of the problem of what one might call cognitive confidence that sank the proposal to ratify the Nice Treaty in 2001.
It is essential, therefore, to assess just how prevalent the feeling of lack of understanding is. The Irish Times/TNS mrbi series of polls and the results of polls conducted on behalf of the European Commission Representation (ECR) in the wake of Nice 1 and Nice 2 provide some help in teasing this out.
At the beginning of the Nice 1 campaign not much more than one-third of voters felt that they understood at least some of the issues involved. And it didn't get any better as the Nice 1 campaign unfolded - see the first and second bars in the accompanying graph.
The central story of the second Nice referendum is that a feeling of having some reasonable knowledge of the issues was much more extensive from the beginning of the campaign - up from 37 per cent in Nice 1 to 53 per cent in Nice 2. And it got better, with 61 per cent of respondents in the ECR post-Nice 2 poll feeling that they knew at least some of the issues involved.
This brings us to now and to the first Irish Times/TNS mrbi poll of the Lisbon campaign. On the same measure of subjectively assessed knowledge of the issues in the referendum, the situation is now worse than it was at the corresponding point in the first Nice referendum campaign. As of the beginning of last week, when fieldwork for this poll was conducted, only 29 per cent felt even moderately comfortable with their level of knowledge of the issues (compared with 37 per cent in Nice 1) and 40 per cent said in effect that they hadn't a clue (compared with 31 per cent who said they didn't know what the treaty was about in Nice 1).
The perilous state of the public's knowledge was confirmed by research conducted by the RedC polling organisation for the Referendum Commission that showed that 5 per cent of voters said they understood it very or quite well, 15 per cent said they understood it to some extent, with, as the commission press release put it, a full 80 per cent saying either that they did not understand it particularly well, or did not understand it at all.
The polling evidence also shows that people's subjective grasp of the issues in the Lisbon Treaty shows up in their voting intention (or lack of intention). Thus, very substantial majorities of those who have a good understanding of the issues and of those who feel they understand some of the issues intend to vote for the treaty (at rates of 66 per cent and 62 per cent respectively - see accompanying table).
Support for the treaty drops to 39 per cent among those who are only vaguely aware of the issues and to a mere 10 per cent among the two-in-five voters who say they do not know what the treaty is about at all.
If the entire latter group were to abstain, the treaty would be passed. However, if they turn out in significant numbers and translate their doubts into a No vote, the outcome will depend on the rate of turnout among the better informed. The fact that 31 per cent of those intending to vote No give lack of understanding or lack of knowledge as their reason is a further warning to the Yes side.
The "if you don't know, vote No" lobby may be smaller than it was in Nice 1 (when it was 39 per cent) but it is still large enough to give the supporters of Lisbon cause to worry.
While lack of knowledge, or lack of confidence in one's knowledge, regarding Lisbon is widespread throughout society, there are pockets where it is concentrated. As one would have expected, it differs with age, with a 13 percentage point gap in levels of good/some knowledge between those under 25 and those aged 50 to 64.
Lack of confidence in one's knowledge is also more prevalent among women, particularly among younger people - the gap between males and females among the under 25s is 16 percentage points, compared with five points in the 50 to 64 age group.
Subjective assessment of knowledge of the treaty is also related to occupation and social class. There is a 25 percentage point gap between levels of subjectively assessed knowledge as between those in upper middle- and middle-class occupations and those in working class occupations.
All this suggests that the Yes information campaign needs to make more strenuous efforts to reach beyond middle-class, middle-aged, males. In assessing the current polling evidence, two aspects must be kept clearly separate. The first is the voting intention figures. Here the message is mixed. Judging by the first opinion poll in Nice 2, the Yes side can win while starting out with only 37 per cent of the vote. Equally, however, the data on Nice 1 shows it can lose while starting out with 52 per cent.
The message here is that voting intention at this stage is volatile and one should not read too much into it. The measure of subjective knowledge presented in the accompanying graph is probably more reliable and the message here is that, in this respect, public opinion now is more like opinion in Nice 1 than opinion in Nice 2. Those who wish to alter this will have to put in great effort.
Prof Richard Sinnott is director of the Public Opinion and Political Behaviour Research programme at UCD. Various related papers can be accessed at http://geary.ucd.ie/popb/