Sun Responds to Polling Bollocks

Cross-posted to the BlogsCanada election blog.

I promise, this is my last entry on polling. For at least a couple of days. On Friday, I disparaged the Vancouver Sun for the deeply-misleading polling “data” on their front page. I sent them an email that basically reiterated the points made in my posting here. Kirk LaPointe, the managing editor, sent the following reply:

Mr. Barefoot:

I am writing in response to your note about our poll results in today’s paper.

As you may have noticed, we’ve worked to de-emphasize poll results in our election coverage. Even though we’ve provided ample information about election polls inside the paper, today was the first day we’ve placed such a story on our front page. Our overwhelming body of work has been on the party platforms.

As for the Compas poll, the 600-person sample is lower than some national polls, but considered statistically reliable in order to derive the story we did.

On the matter of the provincial results, we are carrying a further story Saturday on the NDP result to clarify the inherent shortcomings of drawing many conclusions with such a small sampling.

Thanks for your note.

Kirk LaPointe,
Managing Editor,
Vancouver Sun.

“Considered statistically reliable in order to derive the story we did”? That’s careful journalist-speak, isn’t it? So statistical reliability is relative to the nature of the story? What kind of story can I write if I ask 3 people and a schnauzer who they’re supporing?

Regardless, I must credit the Vancouver Sun with responding to my emails in a prompt and professional manner. On two previous occasions I’ve had heated but gratifying (that sounds kinkier than I intended) email exchanges with Patricia Graham, the editor-in-chief.