Q+A snap poll

Q+A snap poll

Yesterday, Q+A published a Colmar Brunton snap poll, which purported to show that the country was evenly split over whether Judith Collins should stay in Cabinet, but that the issue was not decisive in their vote choice for a large majority.

I'm not at all surprised that most people do not see this as an important issue. What should concern National is that any voters, including apparently 8% of National supporters, would find stories like this important. 8% of National voters is a little under 4% of the population, and the left only needs to pry around half of them loose to give them an electoral edge. Also, if you asked "is the perception that National is out for its own mates - rather than everybody - an important consideration," which is the larger issue Collins' debacle feeds in to, then you might get a more noteworthy response.

Coming back to the first point, there was some chatter yesterday about the quality of the sample. In particular, some were concerned that the poll had too few lefties (153 Labour and Green voters combined, which is 30.6% of the sample), while having too many right wingers (222, or 44.4% of the sample).

Certainly that gap - almost 3:2 - is very large compared with all other recent public polls . Colmar Brunton came on Twitter to say that those numbers were not weighted by likelihood of voting, so I asked them to provide the average likelihood of voting scores across the parties so we could check. They never replied, so it is hard to know for sure, but I think filtering on likelihood to vote would make these raw numbers more, not less, skewed. This is because on average people who prefer the left are less likely to cast a ballot than people who support the right. (That is why high turnout is often seem as good the the left, while rainy election days are seen to favour the right). Unlike most of their polls, this one was conducted almost entirely during the week (90%+ of interviews likely took place during the week), which I've shown before can have an impact on left-right balance.

So can we take anything much from the poll? Not a whole lot. I suppose we can see that the Judith Collins mess has pierced the public's consciousness outside the beltway, because a large majority of people were willing to express a view on it. There's not much upside for National in that.

Filed under: 

Commenting on this Blog entry will be automatically closed on May 16, 2014.

Add new comment