The Presidential Election

Sir, - Does anyone in The Irish Times take an interest in simple arithmetic? You published an MRBI Poll last week

Sir, - Does anyone in The Irish Times take an interest in simple arithmetic? You published an MRBI Poll last week. The poll was up to the usual high professional standard, but ever since then your reporters have been completely misinterpreting it. Has no-one actually read it, or looked at the figures in it?

I know a poll is a snapshot of opinion, as the cliche has it. But your poll showed Adi Roche in second position, and not in third. And there is no information contained in the poll, apart from some second-preference indications, that enable any precise calculations to be made after the second count.

When "don't knows, etc." are excluded from the poll, it shows a sample of 876 voters. Those voters filled in their ballot papers this way:

McAleese: 352 votes; 40% Banotti: 233 votes; 26.6% Roche: 223 votes; 25.5% Scallon-Dana: 68 votes; 7.8%

READ MORE

In other words, after the first count, Banotti and Roche are separated by no more than 10 votes out of 876. That's half the gap you reported.

When Scallon-Dana is eliminated, as she would be on those figures, 59 of her 68 voters expressed a second preference. The MRBI figures are:

McAleese: + 26; 378 votes; 43% Banotti: + 9; 242 votes; 27.6% Roche: + 24; 247 votes; 28.2%

While remaining transfers in the scenario (after the elimination of Banotti) seem to favour McAleese at this stage, it is impossible to draw any hard and fast conclusion, because one would need to know some third-preference figures (because of the prior elimination of Dana), and these were not published by MRBI.

The key thing, from the point of view of inaccurate reporting, is this: your own MRBI poll shows Banotti being eliminated before Roche. Your reporting throughout the week assumes the opposite. Opinion poils have been described in the past - especially when they appear so early in a campaign - as psychological weapons.

What a pity that a false psychology can be generated by inaccurate reporting of simple enough figures. Next time, could someone read them properly? - Yours, etc.,

Fergus Finlay,

Joint Campaign Manager, Adi Roche Campaign, 8-9 Upper Camden Street, Dublin 8.