Polls are for the dogs...

There's more than one problem with this poll:

"The results of a poll released today show that Canadians are rejecting Stephen Harper's plan to distribute a $1,200 child allowance to parents with children under six years of age. The Environics poll asked more than 2,000 Canadians their views on child care and the response was clear - 76% of Canadians support a national affordable child care strategy such as the 2004 federal-provincial agreement that was cancelled by the Conservative government."

"Monica Lysack, Executive Director of the Child Care Advocacy Association of Canada (CCAAC) says that Canadians are not buying into the Harper government's child-care strategy adding, 'They see that this plan isn't going to help them find affordable, quality care for their children.' The poll showed that only 35% of Canadians support the government's child allowance plan."

First of all what are the undecideds? These might just be decided voters, in which case the actual split might be drastically different.

Next where is the 95% confidence interval for the poll? That's usually standard piecemeal givings for any news story.

Also how many people actually knew what in the heck the 2004 agreement was?

What were the questions asked? These are all glaring holes in this poll.

And finally, just how does this match up to other polls? Has support dropped from another point?

All good questions. But they mean nothing. Because another poll was taken early on this year. It was called an ELECTION.

Addendum
------------------------------
Apparently the plot thickens:

"Dryden’s criticism of the Conservatives’ child-care plan also seemed to receive a bit of a boost this week, when an Environics Research poll revealed that 50 per cent of Canadians prefer the national day-care program proposed by the former Liberal government. In comparison, 35 per cent said they favoured the Conservative government’s plan."

What happened to the 70%? Apparently, just like Liberal principles, they are a "lving" concept subject to change.