Estimating the “house effect” for each pollster. How much do they differ from the overall average for each party.June 11th, 2014
All of us who follow the polls closely know that some firms will be particularly beneficial to one party or another and generally produce some of the worst figures for another party.
Now as part of a methodology change the “Polling Observatory” at Manchester University in their latest report has sought to measure this as part of a big change in how they average the polls.
A spin off from this is that they’ve sought to estimate the “house effect” for different firms. This is how they describe it:-
“Our new method makes it possible to estimate the â€˜house effectâ€™ for each polling company for each party, relative to the vote intention figures we would expect from the average pollster. That is, it tells us simply whether the reported vote intention for a given pollster is above or below the industry average. This does not indicate â€˜accuracyâ€™, since there is no election to benchmark the accuracy of the polls against. It could be, in fact, that pollsters at one end of the extreme or the other are giving a more accurate picture of votersâ€™ intentions â€“ but an election is the only real test, and even that is imperfect..”
So the best firm for the Tories is Populus, for Labour YouGov, the LDs ICM and Survation for UKIP.
Flick through the party options on the chart to see how this works out. The biggest variation is for UKIP. ComRes phone gives Farage’s party 2,5% below the average while Survation is 4.4% above.
This is a very useful resource.