Today YouGov released poll data that UKIP are on 14 per cent - something that wasn't news to me as this is exactly the same as the calculations I made last week. However the figure looks decidedly dodgy.
Andrew Spooner is a London-based blogger and writer
Today YouGov released poll data which had Ukip on 14 per cent – something that wasn’t news to me as it is exactly the same as the calculations I made last week.
YouGov’s poll comes after data published by ICM in the Guardian which claims Ukip has an 18 per cent vote share.
So how could Ukip achieve this much of the national vote share? This figure, too, looks decidedly dodgy.
If we take 31.5million or 66 per cent of voters as a likely General Election turn out, Ukip would need 5,670,000 to reach 18 per cent. Of that 31.5million, if Ukip polled a uniform 20 per cent or 3,700,000 across the roughly 18.5million voters who might turn out from the UK’s shires, towns and small and medium cities and combined that with 12.5 per cent or one million from eight million voters in the major cities Ukip would then need 970,000 or 19.4 per cent from Scotland, Northern Ireland and Wales to reach 18 per cent nationally.
Given that Ukip just don’t exist in Scotland, Northern Ireland and Wales, a 19.4 per cent vote share there is not only very unlikely it would be miraculous.
Give Ukip a more likely and still very generous 7.5 per cent in the major cities and five per cent in Scotland, Northern Ireland and Wales, and they would need 4,820,000 or 26 per cent across the rest of the country to make 18 per cent of the total vote share. Again, very unlikely.
Bring Ukip’s vote share down to the 4.16 per cent they achieved in Bristol – the only major city they’ve recently competed for the vote in – for the urban vote, totalling 330,000, and down to the 0.5 per cent or 25,000 vote share they took in Scotland, Northern Ireland and Wales in 2012, Ukip would be left needing 5,312,000, or 28.7 per cent in the rest of the UK, almost a full one per cent above what they polled in Eastleigh, to make the figure up to 18 per cent.
It just doesn’t stack up I’m afraid.
41 Responses to “Ukip’s 18 per cent debunked”
henrytinsley
I don’t understand this article. ICM and YouGov have goodish track records and they’re quite likely to be broadly correct. It’s not unknown for 3rd or 4th parties to have high numbers, especially mid term
johndelancie
Articles like this just smacks of the general out of touch politics the Lib/Con/Lab are peddling to try and play down the UKIP threat. Still the political elite just don’t get it and won’t listen to the electorate. Lib/Con/Lab will get everything they deserve come the EU elections and general election in 2015
Sgt. Analbeadz
Of course, Gandhi also had lots of sex with very young girls. Just saying.
Mathew Blackshaw
Both of you remind me of the Tea Partiers who laughed at Nate Silver and PPP.
Who laughed last?
Sgt. Analbeadz
I think what the author is pointing out is that UKIP are extrapolating a national result from local level data that they know to be unrepresentative of the nation as a whole. Manipulating and misrepresenting statistics – in a really obvious way – and then getting caught out doing it is, like, 90% of everything they do.