Eurovision stats tell Tel's tale of tit-for-tat

On the eve of the annual Eurovision Song Contest , veteran host Terry Wogan will be heartened to learn that his mischievous theory of tit-for-tat voting has been confirmed by academic research.

Alex Haslam of Exeter University's School of Psychology and Bertjan Doosje of Amsterdam University's department of social psychology have investigated Eurovision voting between 1991 and 1996, prior to phone voting by the public. Their study, to be published in the J ournal of Applied Social Psychology , found juries from 21 participating countries indulging in mutual favouritism.

Researchers identified several "cliques" whose jury members favoured performers from other countries in their clique. However, they found no proof that coteries were formed on political, geographical or religious lines. Professor Haslam said: "Although the cliques exerted a statistically strong impact, their existence was denied by contest organisers and cannot be discerned without complex analysis. It is not even obvious that the voting panels themselves were aware of them."

The UK formed a group with Ireland, Sweden, Iceland and Austria. Ireland was also in a group that included Turkey and Bosnia-Herzegovina.

"You could argue that one reason that Ireland did quite well was its being looked on favourably by two clusters of nations instead of one," Professor Haslam said. It was impossible to say whether reciprocal voting was more important than the quality of the music, but it made a difference, he said.

A separate mathematical analysis of votes in the contest from 1992 to 2003 by Oxford University physicists, meanwhile, also confirms the formation of cliques among different countries. Their paper has been submitted for publication to the journal Physica A .