Surveys or opinion polls are commissioned to get to know what people think about a particular product, topic or situation. Scientific surveys carried out indicate the margin of error, and the larger the sample taken, the smaller the margin of error.

The producers of the TVM programme Xarabank are known to commission several surveys, and one of their more common subjects is politics. At least once a year – and to coincide with an edition of the programme that has as guest the Prime Minister or the Opposition Leader, or both – they ask people a set of questions linked to their political allegiances.

The respondents are asked, among others, who they voted for at the last election, who they will vote for at the next, and who of the political leaders they trust most.

The last such survey was published yesterday week and showed a shift in the electorate towards the Malta Labour Party.

Although previous surveys carried out by the same team over the past years had also indicated that the MLP was gaining ground – something that was also confirmed in local elections held since 2004 and the EP election – for the first time there was also a shift in favour of Opposition Leader Alfred Sant.

In the past, Prime Minister Lawrence Gonzi had always prevailed over Dr Sant in terms of trust, even when more people had said that they would have voted Labour. But this time, 37.9 per cent of the respondents preferred Dr Sant against the 30.5 per cent who said Dr Gonzi when asked who they would trust as Prime Minister.

The percentages are similar to those obtained for the two other main questions – 39.5 per cent said they voted Labour in 2003 and 37.9 per cent said they would vote MLP next time round, while 33.9 per cent said they voted PN in 2003 and 29.9 per cent said they will vote PN in the coming election.

This could be interpreted as a strong shift towards the MLP, just before the local elections are held next week and, more importantly, a few months away from the next general election.

But these figures must also be seen in the wider picture – because 25 per cent of the respondents would not say who they voted for in 2003, and an even bigger chunk, 30.5 per cent, would not reveal their preference for the coming election.

In numerical terms, and considering that the number of eligible voters for the next election will be just over 300,000, this would amount to a body of between 90,000 and 100,000 people. This is a large number, and it looks even larger because general elections in Malta have always been decided by a few thousand votes, such is the strength of the two main political parties.

Of course, some of those who did not want to reveal their favourite political party did so because they did not want to, not because they do not know who they are going to vote for. But then there are others who are really undecided, and who make their choice not on party allegiances, but on a genuine analysis of the prevailing situation, without looking at faces or political creed.

These are the people who will be deciding the election. These are the so-called floating voters who do not attend meetings or party activities, but who get both sides of the story before forming an opinion, and at the end of the day will cast their preference in favour of the party they believe will be best for the country.

The major political parties must concentrate on this 30 per cent or so of the electorate, rather than preach to the converted, as very often happens when politicians address the masses.

It is only by securing the votes of this undecided mass of people that one of them will win the election.