Would it have been different if other words had been used?
There’s a poll finding out overnight which seems to suggest that voters have had enough with the media and Tory demands for Brown to say “sorry” for his past mistakes on the economy. Labour is taking great comfort in the survey and, indeed, Nick Palmer MP was posting here at 2am to alert us.
But is the poll all it seems and is the simplified way the PH present the material in the chart above showing a possible distorted picture?
For respondents were NOT, as implied by the chart, given the simple choice of responding “give up” or “keep pressing” when the question was put.
The actual wording on the first option was – “Give up on this and move on to covering more important issues”. That addendum, in my view, could be described as leading. The use of the phrase “more important issues” is not neutral.
If respondents had been presented with the “give up” or “keep pressing” options as in the chart then my guess is that the response might have looked differently.
As regulars will know I’m far from convinced about the value of non-voting intention questions and this is a good example of one of the problems. Trying to find ways of wording them in a form which is not suggesting to the respondent that they should answer in a particular way can be quite challenging.
PH claims that their finding is “likely to give both media editors and opposition politicians pause..:” Maybe. But it would carry a lot more force if it had been done differently.
This is a hot political potato at the moment and pollsters should expect everything about their approach to be subject to scrutiny.
Update – PoliticsHome response. Freddie Sayers of PH has responded to the criticism at post 113.