A look back at the polls: August

by Stephen Tall on September 1, 2008

We tend not to be too poll-obsessed here at LDV – of course we look at them, as do all other politico-geeks, but viewed in isolation no one poll will tell you very much beyond what you want to read into it. Looked at over a reasonable time-span and, if there are enough polls, you can see some trends.

Here, in chronological order, are the results of the seven polls published in August:

Tories 46%, Labour 26%, Lib Dems 17% – YouGov/News of the World (10th Aug)
Tories 45%, Labour 25%, Lib Dems 18% – YouGov/Sunday Times (17th Aug)
Tories 44%, Labour 29%, Lib Dems 19% – ICM/Guardian (19th Aug)
Tories 48%, Labour 24%, Lib Dems 16% – MORI, unpublished (21st Aug)
Tories 48%, Labour 26%, Lib Dems 16% – YouGov/Compass
Tories 46%, Labour 25%, Lib Dems 16% – ComRes/Independent (24th Aug)
Tories 46%, Labour 26%, Lib Dems 16% – YouGov/Telegraph (28th July)

Which gives us an average rating for the parties in August as follows, compared with July’s averages:

Tories 46% (+1%), Labour 26% (n/c), Lib Dems 17% (n/c)

So it’s very much a case of as you were; unsurprisingly, really, as August is rarely a time of political drama – though Labour may perhaps have hoped that some of Team GB’s Olympian stardust might have brushed off on Labour’s own Team GB.

For those of a psephological bent, there were two must-read articles published last week – by Robert Worcester of Mori on PoliticalBetting.com, and Anthony Wells of the UK Polling Report blog – focusing on how accurate the polls are. Bob Worcester defended their accuracy, demonstrating that c.95% of polls are within the +/-3% margin of error statisticians would expect, and anything outside of that is inevitable sample error.

Anthony Wells meanwhile accepted Bob’s figures, but pointed out that the varying methodologies of the five principal polling companies account for much of the difference in the results they generate:

Taking polls since the start of June:
YouGov average is CON 46.4%, LAB 25.8%, LDEM 16.6%
Ipsos MORI average is CON 46.7%, LAB 26.3%, LDEM 16%
ComRes average is CON 45.2%, LAB 25.7%, LDEM 16.8%

All very close, but…
Populus average is CON 43%, LAB 26.7%, LDEM 19%
ICM average is CON 43.8%, LAB 27.4%, LDEM 19%

Part of this is because ICM and Populus are measuring slightly different things from the other pollsters. YouGov, MORI and ComRes’s figures are based on how people say they will vote. ICM and Populus’s also take into account how ICM and Populus think people who say don’t know will vote, and at this moment in time this cuts the Conservative lead. The rest is likely to be down to differences in their respective weightings.

Does all this matter? Well, to the Lib Dems it does. Out of the seven August polls, only one, ICM’s, was conducted by either of the two companies Anthony identifies as being more favourable to the party. Indeed, four of the seven polls were by YouGov, which has long been identified as a pollster whose methodology gives the Lib Dems lower ratings.

In a world in which the political media were statistically-literate, of course, the polls would be taken with a large pinch of salt.