False dichotomies run amok

If Decima's latest poll results are to be believed, 92% of all people planning to vote for the Liberals have absolutely no reason not to vote NDP:

The poll suggests Harper's personal unpopularity helps sustain the Liberal party; almost 40 per cent of people who said they would vote Liberal cited Harper as their main motivation.

Another 52 per cent said they favoured the Liberals because they disagreed generally with Conservative policies.

Mind you, it would have helped if the poll included any criteria which could have helped sort out voters among the various reasons to vote (or not vote) NDP as compared to Liberal. And in keeping with that lack of options, both stated reasons to vote out the Liberals were based more on Liberal action than anything related to the opposition parties:

The Decima survey turns conventional wisdom on its head by suggesting voters' inclination for a change in government dwarfs their anger over sponsorship. Only 35 per cent of respondents who said they wanted to replace the government cited the scandal as their prime motivation.

A far greater number - 57 per cent - said it was because the Liberals have been in power too long and they wanted a turnover after 12 years.

What's even more odd is that the wording of these options suggests that any scandal on issues other than sponsorship itself should be placed in the latter column rather than the former. As a result, even the most-publicized finding from the poll could be based more on the wording of the question than on any genuine measure of the public's views.

Polls are probably overused and overanalyzed as a means of measuring public opinion anyway. But if they're going to be commissioned, it helps if they at least try to approximate the range of options actually available to respondents.