Local Election Opinion Polls
Looking at my referral logs, if you aren't a regular reader of this site the chances are you are here looking for opinion polls on the local election. The bad news is that there is only really one.
Not many pollsters cover local elections since they are trickier to poll than general elections. Firstly not every local authority has local elections today, so a poll has to distinguish between respondents in areas with a local election and those without. Secondly, unlike general elections where all three main parties contest nearly every seat, in a large proportion of council wards this isn't the case. If you are conducting a poll that seeks to actually predict an election result, what do you do with people who want to vote for a party who isn't standing in their ward? Finally local elections have very low turnouts and, since people aren't very good at predicting if they'll vote or not, this makes it difficult to tell which people to include in a prediction.
This year three polls have been published that asked about how people might vote in hypothetical local elections, but only one (by BPIX) was actually set up to measure how people are likely to vote in the areas with elections.
BPIX's published figures were filtered to include only those who said they were certain to vote in the local elections and their figures are apparantly drawn only from those areas that actually do have local elections today. Their figures were:
BPIX/Mail on Sunday: CON 35%, LAB 26%, LDEM 23%, Oth 16%
If BPIX's poll turned out to actually reflect the result, would this be good or bad for the political parties? The shares of the vote announced on the BBC on polling night are not actually true shares - they are notional figures based on a representative selection of key wards. The actual results are available from the LGC Local Elections Centre at Plymouth University. The seats up for grabs today were last contested in 2002 (in the case of London) and 2004 (in the case of the Metropolitan and District councils). The actual shares of the vote then were:
London 2002: CON 34.2%, LAB 34.1%, LDEM 20.6% England 2004: CON 32.1%, LAB 28.4%, LDEM 25.3%
So how do they compare? Unfortunately, because the polls haven't been divided up to cover just London or just the rest of England we don't know for sure. Certainly if BPIX are correct it appears as though the Conservatives have slightly improved their position, and Labour's support has fallen, possibly quite drastically in London. In around 13 or 14 hours we'll have the first real idea.
(There were also two ICM polls, but these both went to people across the whole country, not just in those areas with local elections, and didn't factor in turnout or ICM's usual adjustment for the spiral of silence. Therefore they cannot be taken as a guide to how people will vote in the local elections.)
