How meaningful are responses to very long questions?

We’ve now got the precise wording of the polling question that was read out over the telephone to 751 interviewees in this week’s Populus poll from the Times and which has been seized upon as evidence that there is little backing amongst the public for the government’s core strategy.

“Which of the following statements do you most agree with?

Britain has a severe debt crisis and this government thinks the country’s budget deficit must be dealt with by the time of the next general election in five years time

Britain has a serious debt crisis and the country’s budget deficit should be halved by the time of the next general election in five years time but dealt with completely over ten years

Protecting the vulnerable and keeping unemployment as low as possible should be a bigger priority than reducing the country’s budget deficit. “

Only 22% of those questioned chose the first option while the other two got 37% each.

Yet in the same poll there were the apparently contradictory findings that 53% approved of the coalition’s policy compared to 45% who disapproved and that 64% blamed the last Labour government for the cuts.

The question above is at 94 words far too long to be asked in a telephone poll and would probably have taken the interviewers nearly half a minute to read out. The experience where you have such a wordy approach is that the earlier points are less likely to be favoured. Interviewees simply forget.

Add onto the the fact the differences in the way the three points were phrased and I don’t think you can read too much into it.

This is, I believe, the first Populus Poll for the Times since its veteran political commentator, Peter Riddell, retired. He was central to the paper’s coverage of polling and I don’t think that he would have let this question through in this form.

Mike Smithson