Now that's how I remember it.

Harris-Decima's got a new poll out. Thought I'd look at The NPD in La Belle Province. Accepting that there are standard preferential differences between polls (and thus they're not directly comparable), we can see that their numbers are much closer to business-as-usual than the Angus Reid poll I recently discussed.


What do they look like? Well, here's the current Harris-Decima numbers in Québec:

• BQ: 37%
• Liberals: 28%
• Conservatives: 15%
• Greens: 10%
• NPD: 9%

These numbers are more than a little different from those Angus Reid numbers, aren't they? They were, with the deviation in parentheses:

• BQ: 37% (even)
• Liberals: 20% (up 8%)
• NDP: 18% (down 9%)
• Conservatives: 16% (down 1%)
• Greens: 7% (up 3%)

So while you can expect a percent or two difference from one polling house to another, can one polling house really say that one party's support is twice what another polling house suggests? Well, apparently so.

Related articles by Zemanta

• Summer tour lifts Liberals, not Ignatieff: poll (thestar.com)
• Support for Tories down across most of country: poll (ctv.ca)
• Masse, Comartin and other NDP MPs could decide fate of long-gun registry (windsorstar.com)
• You've come a long way, Iggy. Is it far enough? (theglobeandmail.com)