Does UK Polling Industry Require More Regulation?

This post is also available in: Español (Spanish) Français (French)

Contributed by Johnny Heald (ORB Research)

The House of Lords enquiry sought to determine, among other things, whether the UK polling industry requires more regulation. I argued that in the UK I believe as an industry we are already regulated. The British Polling Council (BPC) was established following the 2005 election and has strict rules on transparency. As a practitioner we must post data tables within 48 hours of any release, we must be transparent with our methodology and weighting procedures and we have a body which can (and does) independently investigate poor question design. I’m not too sure that there are many other countries with rules in place such as these. There was talk about a French style “commission des sondages” but having spoken with colleagues at BVA France, I’m not too sure how effective this would be. The sheer volume of polls conducted now in the UK on a daily basis would require an army of qualified experts checking each poll, methodology and analysis. There was also some talk of a ban in the run up to election day but once again I am not sure how this would be policed – what would stop an Irish newspaper or a US website (e.g., Huffington Post) collecting and releasing data during the ban? This is certainly the case in France where French speaking newspapers in Switzerland conduct polls in France and hashtags such as #radiolondres trend heavily during the ban.

Experts have argued that the BPC could and should do more. The BPC is a voluntary organisation which is funded by the pollsters. There is perhaps some merit in making it look a little less like a cartel (so, for example, it could take a representative at the board level from WAPOR, or the Market Research Society, or perhaps from Chartered Institute of Journalists) while there is also a strong case for it to also play more of role in educating the public (and media) in how to report polls. I would also argue that there is possibly more oversight required on the issue of question wording because as practitioners we are all called from time to time by clients who have an agenda and want to write questions to get answers which advocate their cause. This is an area where the industry does, from time to time in my view, let itself down.

As Claire Durand points out in her article, one area where pollsters are dangerously heading towards is the art of prediction.  When I studied politics at University and when my father (who was managing director of Gallup poll in the UK for 15 years) taught me about polling one of the first things he said was, “This is a snapshot of public opinion collected over the previous week; this is not a prediction”. Yet we now operate in a world where we are being asked to give the percentage change of something happening or how many seats will change hands next week, etc. While some ‘predictions’ have been highly accurate others have also been a long way off. The problem we have is that the financial markets also react very much on polling data where the race is tight and the potential result significant – the Scottish referendum and Brexit vote are two examples. Poorly reported polls (Scottish referendum) and poor sampling (Brexit), both resulted in millions of pounds being wiped off the value of sterling in one days trading – all for a handful of questions which cost less than £500 to ask and the change from the previous poll also within the margin of error.

Finally, there has for a long time been a debate about the actual influence polls have on voters in elections. Based on my 25 years working in the industry I disagree with any suggestion that the polls actually influence vote. Sure, they undeniably steer campaigns. If the Comey FBI influence in the US 2016 election outcome is still being debated, I struggle to believe that polls actually influence vote in the UK. Having conducted more than 200 political focus groups up and down our country not one person has ever said ‘polling’ when answering the question, “What are the three main issues which will impact how you are going to vote in the upcoming election?”