Why did the phone pollsters over-state Labour?
Following last nights results the pollster Ipsos-MORI has announced that it will be carrying out a review of its approach and the phone firms generally asking “why empirically, they appear to have tended to over-estimate the Labour share of the vote”.
An issue for the firm, of course, was that its final poll was completed a week before the election and a lot can happen in the final few days. That’s certainly true but there was a disparity between its figures and those of the online pollster, YouGov from February onwards when details of the first MORI survey were made public.
The results from MORI, MRUK and ICM have been far better than previous phone polls ahead of the 2000 and 2004 races. The big difference this time is that the contest was far closer.
To my mind the difference between completing an on-line questionnaire and responding to an unsolicited randomised phone call is that with the former you decide when you want to respond. The nature of the phone survey is that you have less control of the timing and this, I would suggest, might lead to different sorts of people taking part.
The internet responder is being proactive while a telephone responder is passive.
This factor has become increasingly important as in this online age we use the internet for so much more. The user is in the driving seat so much more of the time.
In comment on UK Polling Report Ben Page of MORI writes: “…we will be reviewing our political polling methodology to look at lessons learned and publishing our conclusions and any changes we make in a few monthsâ€™ time. In my opinion thatâ€™s the best thing to do. There is nothing to be gained from just blaming a result that looks wrong on a late swing, the right thing to do is to look in detail at the figures, try and work out what if anything did go wrong, and see how it can be put right.”
We will follow this with interest.