The Green Party has sounded a "serious health warning" on a poll of how people would be likely to vote in the general election in Brighton, calling it "way out" and "substandard and misleading".

A telephone poll of 336 voters in Brighton Pavilion constituency commissioned by The Argus local newspaper gives Labour a surprising and incongruous lead with the Green Party in third place behind the Conservatives.[1]

This is at variance with all other polling and electoral information, including a 533 voter poll conducted by ICM Research, a reputable national polling company and member of the British polling council, in December.[2]

It is also inconsistent with the prevailing political mood of the electorate and local and European elections held over the last two years and the general election of 2005 when the Greens polled 22%.

The Green Party says the poll by Kindle Research does not conform to the standard professional practice for voter intention surveys undertaken by reputable political polling companies that are members of the British Polling Council including Ipsos MORI, ICM and YouGov.

Chris Rose, National Election Agent for the Green Party, said, "Whichever way you look at it, this poll is completely substandard and misleading.

"While all polls have a margin of error, if a newspaper wants to commission a professional survey of local voter intentions, they should commission a poll of at least the standard sample size of 500 voters with a question structure that conforms to British Polling Council member standard practice.

"If this poll were a packet of cigarettes the 'skull and cross bones' health warning would need to cover the whole packet.

"It's the nearest thing in the research industry to a polling health hazard."

Paul Steedman, National Campaign Director for the Green Party, added, "We expected a few voodoo polls in this highly charged campaign and this is the first one but I suspect it may not be the last. The poll is so way out, topsy-turvy and against all the evidence we have from our doorstep campaigning that it might as well have been commissioned by Alice in Wonderland.

"It might make an interesting topic of conversation in the pub but it is hopeless and flawed research. The hard facts suggest this is a two-horse race between the Greens and Tories.

"Our campaign has already achieved a critical momentum as voters are hearing our message and we step up the fight to stop the Tories."

Greens point to the following weaknesses in the research base of the poll:

Unclear professional credentials for political polling: Kindle Research does not appear to have any significant background in political polling.

Failure to screen out non-voters: the Kindle Research poll failed to follow usual practice by screening people who said they would not vote.

Non-voters are irrelevant to the outcome and should be screened out via preparatory questioning on likelihood of voting.

Failure to prompt parties as is standard practice by BPC member companies: there was no prompt for any party - which is not usual practice in BPC member political research. Respondents in voter intention opinion polls are usually prompted for the main challenging parties in a contest.

This is because respondents may not have considered the choices in a telephone poll fully in the way they would do so in a real polling situation. ICM Research have suggested this poll does not appear to follow British Polling Council guidelines and does not conform to their own professional practice for undertaking voter intention research. Voter intention polls follow stringent guidelines, and are carefully formulated to be as accurate as possible, including prompting the names of political parties. Failure to do this automatically biases the result towards the larger parties.

Failure to use a standard size voter sample: the sample of voters is too small to be reliable. The poll sample was only 336 people in Brighton Pavilion, meaning the margin of error on a professional poll is approximately 10% rendering this poll, with all its other weaknesses listed in this statement, too unclear to be meaningful.

Incongruous result with other election predictions: most leading bookmakers have Green candidate Caroline Lucas as favourite to win Brighton Pavilion seat. The Kindle Research poll prediction suggests Labour also lead in the hyper marginal neighbouring constituencies of Hove and Kemptown. This differs to the mood on the doorstep with most pollsters currently predicting Conservative gains there.

For more information please contact Brighton and Hove Green Party office on 01273 766 670. 

Notes

1. The Kindle Research-Argus poll puts Labour on 26%, the Tories on 16%, and Greens on 12%, with the LibDems on 5% and others 2%.

2. A recent ICM poll puts Greens ahead of the Conservatives and Labour - with Greens on 35%, the Conservatives on 27% and Labour on 25%. This is congruous with YouGov study for politicshome.co.uk, which has predicted a Green win.**

3. The European elections in 2009 saw Greens take the largest percentage of the vote in Brighton, gaining 31%, ahead of Conservatives (22%) and Labour (15%). Council elections also support a growth in Green support. Greens have the same number of councillors sitting on Brighton and Hove City Council as Labour and in a recent by-election the Greens took a seat from the Conservatives.

**December poll - ICM Research interviewed a random sample of 533 adults aged 18+ by telephone on 16-21st December 2009. Interviews were conducted across the political constituency of Brighton Pavilion and the results have been weighted to the profile of all Brighton Pavilion adults. ICM is a member of the British Polling Council and abides by its rules.
_______