The problems of polling the electorate

A very interesting report from the Polling Observatory which looks at recent trends in voting intentions: How to read the polling tea leaves

In particular there is a helpful discussion of 'house effects', i.e., the tendency of a polling company to report high or low figures for a particular party. However,if a pollster tends to show one of the parties doing better than the polling industry on average, it does not automatically mean their estimate for the other main party will be lower than the average.

Prompting for the Brexit Party and controlling for past vote appear currently to have significant impacts on poll numbers. In the former case, pollsters that prompt for the Brexit Party in their surveys tend, unsurprisingly, to report higher numbers for the party.

The use of past vote (i.e. how people voted in 2017) to weight samples to make them representative is a longstanding practice in the polling industry. However, this can introduce error through people misreporting their past vote, leading supporters of a party to be overrepresented in the poll.

They wisely conclude, ' There can be no way of knowing which pollster is right before election day, but it is worth urging some caution in how these sorts of numbers are interpreted by those in politics, media and the wider public.'

YouGov's shock poll

A poll from YouGov published today based on constituency by constituency estimates suggests that there could be a hung Parliament.

Yet Stephen Fisher of Oxford University predicts a Conservative majority of 100. A ICM poll released yesterday records a 12 per cent Conservative lead and implies a majority of 76. The spread betting company IG Index suggests a Conservative majority of 106.

It should be noted that YouGov themselves make the caveat that 'it would only take a slight fall in Labour share and a slight increase in the Conservatives' to result in Mrs. May returning to No.10 with a healthy majority.'

It is also stated that 'The model is based on the fact that people with similar characteristics tend to vote similarly - but not identically - regardless of where they live.' This strikes me as a big assumption given that these characteristics are a less good guide to voting behaviour than in the past. It is also admitted that the model 'does not account for specific local factors that may shape the vote in some seats.' This strikes me as important in this election, e.g., popular Labour incumbents.

All the polls face the problem of estimating turnout. For example, 18-24 year olds overwhelmingly back Labour, but will they turn out and vote?

It is well worth looking at the latest report from the Polling Observatory, although they point out that current forecasts could mean anything between a Conservative landslide and the Tories scraping home. One point they make is that Labour may pile up votes in its safe seats, while losing ground in marginal: Polling Observatory

SNP local elections victory looks less secure

Leading elections expert takes a close look at the SNP's performance in the local elections and says that it looks less secure when you examine it closely: Starts to crack when you look closely

Millwall fans to contest general election

Millwall fans are planning to nominate a candidate in one of the Lewisham constituencies in the general election: Protest candidate

They fell out with the local Labour controlled council and mayor after an attempt to use a compulsory purchase order to acquire some of the club's land.

72-year old local resident Willow Winston is to stand in Lewisham East, the leafier part of the borough: Everyone hates us, we don't care