View my bio

Now the Details

Media, ethics, and journalism. What works. What doesn't.

Jeffrey Dvorkin

Wednesday, September 19, 2012

Pollsters and Journalists ARE Biased - in favor of Change!

Dana Millbank writing in today's Washington Post expresses the hope that Mitt Romney wins in November because the gaffe-prone Republican candidate makes better copy than President Obama who "speaks in paragraphs."

Toronto journalists following the gaffe-enhanced mayor of Toronto, Rob Ford while on an official visit to Chicago has also not disappointed the attendant hacks.

According to the Globe and Mail, "when one woman told him she’d been to Canada, (Ford) asked where. She said “the part across from Detroit on the river.” “Oh, Manitoba!” the mayor said. “Oh, you were in Manitoba. Winnipeg?” When a reporter explained that she meant Windsor, he replied: “Windsor. So that’s in Ontario..."

Four more years!

Pollsters are also similarly challenged when it comes to dealing with politicians and elections.

In Alberta, some wishful thinking about an emergent third party, the (so-called) Wildrose Party was seen by pollsters as the next government and poised to oust the longtime Conservatives. It didn't happen.

Similarly in Quebec, pollsters anticipated a surge from the Coalition Avenir Quebec (CAQ and yes, it means the same fecal matter in French as in English and Yiddish). That didn't happen either and the Parti Quebecois squeaked in power with a minority government barely edging out the Liberals.

In the US campaign, the polls keep telling us that Obama and Romney are neck-and-neck with the president now starting to move ahead thanks to one of the most inept campaigns ever run by the Republicans. Obviously, the Beatles were right: money can't buy you love. Or votes.

Part of the problem is how to decide on the undecideds. Polling firms in the past have allocated the undecided vote numbers proportionately to the decided categories. This is proving to be wrong.

Undecideds are more likely to stick with incumbents on the principle of the "devil you know."
Undecideds tend to be more conservative but that doesn't always mean they will vote conservative. They may just be unwilling to switch horses. It would be interesting to see how many people actually lie to pollsters. My guess is that it's more than we know...

But pollsters and journalists want a good story (who doesn't?) and often they tend to exaggerate the closeness of the race. It sells more papers.

In fact, the real story can be found outside the newsroom echo chamber and polling offices. It requires shoe-leather reporting but in these times of financial restraint, editors are unwilling to let their reporters off the leash to wander at will.

And the audiences are the real losers.

No comments:

Post a Comment