Yes, Prime Minister was right – polls can be manipulated

Sir Humphrey Appleby was right all along: polling can be manipulated to suit the political objectives of those who write the questions.

In a famous episode of the 1980s sitcom Yes, Prime Minister, the suave Whitehall mandarin explained how polling questions could be written to ensure the “right” answer was received.

The civil servant, played by Nigel Hawthorne, said pollsters could use a series of leading questions to persuade the public to give contradictory answers on whether they supported the reintroduction of national service.

Now a major polling firm has proved that Sir Humphrey was correct.

Ipsos asked the exact same leading questions to two groups of people. One group came down in favour of national service, the other group did not.

The firm said that it would never carry out polling in this way – but it showed the importance of how the wording of questions and the order they are asked can change the result.

The famous scene was broadcast in 1986.

Sir Humphrey, the comedy’s Cabinet secretary, is horrified to hear Jim Hacker, the prime minister played by Paul Eddington, is planning to bring back national service after seeing polling showing it was popular.

He tells Bernard Woolley, the prime minister’s principal private secretary, played by Derek Fowlds, to commission another poll showing the opposite.

Sir Humphrey goes on to explain that this is possible because people do not want to appear stupid in front of those carrying out the surveys.

He said it would be possible to preface a question on national service with four or five other questions which would lead a person to come up with two different answers.

The pollster could then publish only the last answer and not the answers to the leading questions.

To test this, Ipsos asked the same questions used on Yes, Prime Minister – and found that one group was in favour of national service and one group was opposed.

Sample A was asked first: “Are you worried about the number of young people without jobs?”, to which most answered yes.

Then they were asked about whether they were worried about the rise in crime among teenagers, whether there was a lack of discipline in comprehensive schools, and whether young people would welcome authority and leadership in their lives – all of which received a strong yes vote.

Asked finally whether they would be in favour of reintroducing national service in Britain, 45 per cent said yes and 38 per cent said no.

Sample B were asked a different set of leading questions, beginning with “Are you worried about the danger of war?”, to which most answered yes.

The next questions were about concerns over the growth of armaments in the world, whether there was a danger of giving young people guns and teaching them to kill, and whether it was wrong to force people to take up arms against their will.

When asked whether they would oppose the reintroduction of national service, 48 per cent said yes and 34 per cent no.

Each sample consisted of 1,079 British adults, surveyed online at the beginning of February.

National service was introduced in 1939 as the Second World War drew near, and was retained until 1960.

Keiran Padley, a pollster at the firm, said: “What does it all mean? Well, obviously we wouldn’t ever do this poll for real; the wording is too leading and there are rules about transparency of question wording and ordering.

“But it does show the importance of question wording, watching the detail and looking at more than just one poll when you want to understand what the public really think on an issue. Especially in an election year.”

The firm pointed out that Sir Humphrey’s allegation on the show that polling firms can simply decline to publish some of their questions was not allowed under British Polling Council rules.

Yes Minister, in which Jim Hacker is the minister of administrative affairs, ran from 1980 to 1984, and its sequel Yes, Prime Minister from 1986 to 1988.