When reading through today's edition of this newspaper whether online or otherwise, we will be waiting for the first indication of the general election result. The large political parties are sufficiently organised through sampling of the first batches of votes from each ballot box that they can give an early indication of the result. If they take considerable time to announce the first indications, this generally signifies that the result would be close or else that it varies substantially from what was expected.

Over the past weeks we have been inundated with opinion polls seeking to identify trends, changing voting patterns and predicting electoral results. The issue with polls during election campaigns is whether these influence the result rather than simply forecasting it. Specifically, are opinion polls responsible for voter suppression when they predict an electoral defeat of a specific party? Would such a prediction increase the scale of the defeat as voters would have less of a motivation to vote given the anticipated result? 

This argument obviously has an alternative application as voters of parties projected as winners may also conclude that the electoral result is a foregone conclusion and consequently there is no need for them to proceed to the polling station on election day.

Given their potential to influence voters, should the publication of opinion polls be banned? There are various conflicting opinions on this. Those against a ban generally maintain that voters are influenced by many things, including the opinion of others. So, there is nothing new in using the influence argument.

A ban on the publication of polls would not stop the organisation of polls. It would just restrict access to the resulting information and conclusions to those who can pay for it, that is to say political parties and significant lobbies. If the publication of the resulting information is banned it would undoubtedly be substituted by rumours which, as we know, travel fast, and may be more damaging than the forecast results.

In any case banning the publication of opinion polls could well be deemed to be an infringement of fundamental human rights, specifically the right to freedom of expression, a basic right enshrined in our Constitution as well as in the European Convention of Human Rights.

Given all this I think that there is no reasonable case to make for a ban on the publication of opinion polls. However, the information supplied when opinion polls are published should go beyond numbers and explain the background work carried out, thereby explaining the extent to which the conclusions are reliable or not.

Specifically, the number of calls made in order to arrive at the final sample size may be of significant importance, in particular if this number is a substantial multiple of the sample size. This could have an impact on the random characteristics of the sample. Likewise, the spread of the days over which the poll is carried out could be quite significant. A poll carried out over a whole week could definitely be less accurate or significant that one carried out over a couple of days.

So, rather than having opinion polls banned it would be more appropriate if the methodology used is explained in more detail. Transparency is essential, even in opinion polls.

An architect and civil engineer, the author is a former Chairperson of ADPD-The Green Party in Malta.  [email protected] ,   http://carmelcacopardo.wordpress.com