Luntz, as many of you will have seen by now, has been in the employ of RT� for some months, organising televised focus groups of undecided floating voters in order to give us a flavour of what the populace might be thinking in advance of the general election.

His third, and final, attempt to convince us that he is not a political showman will be on The Week In Politics tonight and once again RT� is promising us fascinating results.

Whether those results are any more reliable than they were in the first two programmes, the national broadcaster does not say.

The decision by RT� to fire wads of licence-fee money at a partisan, manipulative and cynical US conservative raises a number of questions.

Could it not have found a partisan, manipulative and cynical conservative of its own? Alternatively, could it not have located somebody suitably telegenic from one of our homebased polling companies, which seem to be proliferating like MRSA at an Irish hospital?

Was it not worried about Luntz's reputation in the United States, where he has been anything but the disinterested moderator he presents himself as on The Week In Politics? Was it at all tempted to drop tonight's programme after the "findings" of Luntz's first show were significantly at odds with subsequent opinion-poll findings, and after the credibility of the second was shot to pieces by the presence in the focus group of one current Fianna F�il member and one former member?

Luntz doesn't deal in facts, figures or statistics. His is a much more touchy-feely approach, in which he tries to tap into into the innermost feelings of people about particular issues and personalities.

The problem with this approach is that it is he alone who interprets everything, he alone who manipulates everything.

Every question he asks, or topic he raises, or debate he provokes is the result of his personal preference;

every answer he receives is interpreted and refined through the prism of his own personal biases.

Instead of being an objective hand who stops the debate from getting out of control, he leads it in particular directions.

This was particularly noticeable in the first programme when he as good as led the focus group towards an extremely warm reception for the idea of a Fianna F�il/Labour government. A few weeks later, an Irish Times opinion poll showed that the electorate was about as excited by a Rabbitte/Ahern alliance as it is by bad weather in April.

Why Luntz latched onto FF/Labour as a winning combination is not clear but, as you'd expect from a US conservative, he is sweet on Bertie Ahern. Only last weekend, he was assuring us, on the basis of no empirical research at all, that Ahern would lead the next government.

This is the objectivity on which RT� prides itself? What is the national station thinking of?

Luntz's partisan, undocumented approach to his work has got him into trouble on many occasions. In 1997, he was "formally reprimanded" by the American Association for Public Opinion Research arising from work he had done for the Republican party on its "Contract With America" campaign. Luntz claimed, as a supposedly objective pollster, that everything in the contract had the support of at least 60% of the general public. When he was asked to provide details of his research, he refused, on the basis that it was confidential.

He was asked for only very basic details, according to Diane Colasanto, who was president of the association when Luntz was reprimanded.

Amongst them, she subsequently told salon. com, were: "How many people did you question? What were the questions? He did finally give us some information, but it wasn't enough. It didn't really explain what the figures were based on. All we could tell was it seemed like there might have been some survey done.

We understand the need for confidentiality, but once a pollster makes results public, the information needs to be public. People need to be able to evaluate whether it was sound research."

Luntz has been a key player in the Republican Party's electoral successes over the last decades. He comes armed with the biases and prejudices you would expect from such a person, and holds views - some cynical, some heartfelt - on topics that are also the subject of heated debate in Ireland currently.

On immigration, for example, he last year wrote a memo advising Republican candidates how to talk about immigration.

"Let's talk about the facts about illegal immigrants, " he wrote. "They do commit crimes. They are more likely to drive uninsured. More likely to clog up hospital waiting rooms.

More likely to be involved in antisocial behavior because they have learned that breaking the law brings more benefit to them by abiding it."

However, there are votes to be got amongst immigrant communities.

So here is how Luntz advised Republican candidates not to speak to Hispanics.

They should not say that "illegal immigrants operate outside the law".

They should not say that they are "a part of an underground community and underground society" or that they "are more likely to commit crimes. . . More likely to drive uninsired. . . More likely to clog up hospital waiting rooms. More likely to be involved in anti-social behavior because they have learned that breaking the law brings more benefits than abiding by it."

Again I ask: was RT� not even remotely concerned about this man's objectivity and methods when they decided to employ him?

Leaving aside the question of whether Luntz is a suitable moderator of an Irish focus group, there must also be doubts over whether such groups are of any use at all.

"You cannot generalise from the results of a focus group, " the American pollster Warren Mitofsky once said. "You can get results you can explore in a real survey but that's all.

A lot of people do research on the cheap and that's a good way to get in trouble."

Which is where The Week In Politics comes in.