Rogue Polling in Brighton and Hove

February 20, 2010 · Filed under Brighton

Since May 2007 the Greens have topped the poll in every election in Brighton Pavilion including City Council elections (May 2007), the Regency ward by-election (November 2007), the Euro Elections across Brighton and Hove (June 2009) and in the most recent poll taken in the Brighton Pavilion constituency (December 2009). During this time, the Green vote reached an all time high of 35% in last December’s ICM poll. (You can read Jason Kitcat’s wonderful graphs here).

In contrast, Labour  have haemorrhaged votes since the 2005 general election when they polled 35.4%. In the European Elections in June 2009, Labour were at their lowest ebb, polling just 14.5% of the city-wide vote in Brighton and Hove.

I mention this because earlier this evening I received, from an unnamed source (I wish I could say whom), a poll conducted by Kindle Research, established in 2005 by Paul Hutchings, a man of no real previous experience in the polling or research of voter intentions. It is one of the most bizarre and suspicious polls I have ever seen, and most definitely an example of rogue polling.

The poll, to be published in the Brighton Argus newspaper this weekend, will cause heads to turn from respected pollsters, of whom I would be interested to hear feedback.

Kindle Research sampled 1000 adults living across Pavilion, Kemptown and Hove. This translates to a sample of around 333 per constituency (actually 336 in Pavilion), exactly 200 less interviewees than December’s ICM poll for Pavilion, thus meaning a greater margin of error in Kindle’s poll to ICM’s.

At no point, in any of the questioning, was a prompt made for any party, neither was any mention made of Pavilion, Kemptown or Hove constituencies. Kindle Research, in their infinite polling glory, asked this: “Which party would you vote for if there was a general election tomorrow?” Is it just me or does this seem a simplistic line of questioning for this poll, given the exceptional context of Brighton Pavilion? Would this not abstract from everything that’s characteristic of Brighton Pavilion and it’s calibre of candidates? I would be interested to know the thoughts of respected pollsters such as Anthony Wells, Andrew Hawkins, Nick Sparrow and Mike Smithson.

According to Kindle’s poll, in Brighton Pavilion, if the general election was held tomorrow, 26% of voters would vote Labour, 16% would vote for the Conservatives, 12% for the Greens, 5% for the Liberal Democrat, 1% said UKIP, 1% said ‘other’, 11% said they would not vote, 19% were undecided and 7% refused. So, a huge 37 % of those that took part in the poll refused to even pin their mast to one of the political parties. Can this be the outcome of well conducted poll and a telling sample? Can these figures be taken seriously, especially off the back of previous polling figures in Brighton Pavilion from the European Elections and the recent ICM poll? Once again, very suspicious stuff.

But, it gets even more worrying.

On page six of Kindle’s ‘findings’, it becomes perfectly clear for whom Paul Hutchings is drawing his analysis for, implying that Pavilion, Kemp Town and Hove, are all straight fights between Labour and the Conservatives. The interpretation of the poll is most certainly an attack on the Greens. No attempt whatsoever is made to make the poll seem credible, certainly not credible enough to be used on electoral literature.

This poll provokes several questions, and answers are needed-

1. How did the poll come about and who commissioned it?

2. What is Paul Hutchings relationship with the editor of the Argus?

3. Why is Kimble Research who, by all appearances look like a market research outfit, carrying out public opinion research, three months before a General Election and in three hotly contested constituencies?

4.  Does this poll bring into question research undertaken by those who adhere to none of the standards of the British Polling Council?

5. Why is the Argus publishing a poll from a research outfit who adhere to none of the British Polling Council’s standards? Do they not care about their own reputation as a credible news source for the people of Brighton and Hove?

Respected pollsters should be worried by rogue polling from researchers who have little or no experience in polling on voting intentions. Paul Hutching’s interpretation of the results suggests a bias in favour of Labour and a pathetic indifference to the Greens. Somewhat surprisingly, he makes no attempt to try and understand why his own poll differed so greatly with that from ICM in December and previous results in Brighton and Hove, e.g. the returns from the European Elections last June.

If this type of poll is aimed to be used on election material, without scrutiny from professional pollsters, it could set a precedent that any type of poll, if not adhering to the standards of the British Polling Council, is an acceptable campaign tool.

Everyone, including all the candidates in Brighton Pavilion, Brighton Kemptown and Hove, would be wise to steer clear of using the Kindle Research polling data in their election campaign. Rogue polling does not deserve any air of credibility, it simply undermines the rigorous,  academic and high-standard polling from the likes of YouGov, ComRes and Populus.