The cost of concealment: why polling needs transparency
3 min read
Public opinion is one of the great shaping forces in British political life, and polling is how we listen to it.
It measures support for parties and much else besides, from attitudes to tax and immigration to what consumers expect of the companies they buy from. Government, business, the press and our own profession all run on it.
Done properly, polling gives the public an accurate account of what their fellow citizens think and gives decision-makers evidence not selected to flatter them. It has earned its standing over decades by showing its working, and the first thing it shows is who asked.
The British Polling Council (BPC) requires any published findings from a social or political poll to name the client who commissioned it. The rule is short and uncomplicated. It exists because a reader can only weigh a poll properly if they know the interests behind it.
Recent days have shown what happens when they cannot. Three polls commissioned on behalf of Reform UK and conducted by JL Partners appeared in the Times and the Telegraph with no mention of the party. The BPC has said its rule was clearly breached, and it and the Market Research Society have opened inquiries. Whether the money behind the polls should have been declared elsewhere is a question for others.
None of this means research must be presented blandly. Polling is routinely commissioned to generate coverage, and that is legitimate. Nobody has suggested the polling in question was inaccurate, and a commissioning organisation is entitled to lead with its strongest finding and frame it for the audience it wants to reach. Selective emphasis is the ordinary craft of communications. Concealment is where it goes wrong.
Yet good practice is not onerous. Name yourself as the client when the findings are published, put the fieldwork dates, sample size and method alongside the story, and agree all of it before fieldwork begins. Where someone else is paying, name them too.
This happened in a country that has already run out of trust to spend. The British Social Attitudes survey finds 12 per cent of people trust governments to put the country before their party, the lowest figure it has recorded. The Ipsos Veracity Index has trust in politicians at its lowest in four decades. A generation of scandals has taught us that the public will not be told where money meets influence.
Polling has kept its authority through all of that because it is one of the few forms of information people treat as a measure rather than a message. A poll on a front page is read as the country's own voice, which is exactly why a party would want its research in print without its name on it.
The damage goes further than one poll or one party. A reader misled by a hidden client is in the dark about more than the provenance of that poll. The instrument they use to weigh every other political claim has been tampered with. Once people learn that a front-page figure may have been paid for by the party it flatters, they discount the honest ones alongside it. The pollsters who follow the rules pay the price for the ones who do not, and the public loses the one honest reading it had of its own opinion. A country that cannot trust its polls cannot know what it thinks, and a country that does not know what it thinks can be told anything.
This is how trust erodes. Rarely through a single large lie, more often through a death by a thousand small cuts, each teaching the public that what they are shown may not be what it claims to be.
Alastair McCapra is CEO of the Chartered Institute of Public Relations
