Could there be a repeat of the 2006 private polling farce?

Word has reached me that there is a possibility that the internet pollster, YouGov might be commissioned to carry out a survey of Lib Dem members on the Clegg-Huhne contest.

Although previous YouGov surveys of Lib Dem and Labour members have achieved nothing like the accuracy of the firm’s Tory leadership polls they could provide a reasonable pointer to those planning to bet. Would they show that Clegg is worth his 1/3 odds on favourite status?

The problem is that Lib Dem elections attract nothing like the media interest of the main parties and the chances of papers commissioning polls might not be that high. So if we are to get information then we might have to rely on privately commissioned surveys which can present problems.

With private polls we are only likely to see the results if the commissioning camp thinks it is in their candidate’s interests and then we are likely to be provided with only restricted information.

Unlike normal opinion polls that appear in the media there is no requirement under the transparency rules, to which most of the leading pollsters subscribe, for private polling data to be published. It is only if it is put into the public domain then we get limited access.

Back in the 2006 race rumours started to develop that a poll, said to have been commissioned by a wealthy Campbell backer, was showing the then 64 year-old ahead, but by nothing like the margin that his supporters must have hoped for. The unknown Huhne was said to be in a respectable second place and this was why, it was suggested, that the full details were not being released.

Then another private poll was commissioned, this time paid for by a Huhne backer, who immediately authorised that the results be published, as they showed that his man was ahead by 4%.

When the full details of this survey were made available people started to question the way it had been carried out. For, unlike normal polling surveys, the question of which of the candidates the respondents would vote for was asked at the end and not the start. It is standard practice with almost all polling to ask the voting intention question right at the start to ensure that the process of questioning does not influence the outcome.

The survey asked people if they thought “Sir Menzies Campbell was too old to be leader of the Liberal Democrats”. It then asked which candidate would most appeal to women, and to “attract young people to vote for the Liberal Democrats”. Another question asked if people thought that “Simon Hughes was unreliable”.

All of this, it was argued, had highlighted the negatives about Huhne’s opponents, so that by the time they got round to answering the voting intention question they might have been more positive about him.

Then details of the earlier private poll, showing Campbell ahead, were made public and precisely the same approach had been taken. The pre-voting intention questions had put the emphasis on Campbell’s positive attributes.

So if we do get private polls then we must press to see the full detail and if you are risking money you should be very careful.

Mike Smithson