The N&O’s Poll Hole
There is a theory about that The News & Observer/WRAL pre-election poll on the
The explanation by the pollster – Del Ali of Research 2000 – was that the poll was accurate when it was taken two weeks before the election. But voters’ opinions changed.
I don’t buy it.
The same outfit had a poll several months ago showing the same result.
Plus, my experience is that voters who say they’ll vote against a bond issue don’t change their minds and then vote for it.
Two local polls – by Public Policy Polling and by the Home Builders of
There’s one question I’d ask the N&O about its poll: How much did you spend on it?
Too often, media outlets buy polls on the cheap. And they get what they pay for. They’re more interested in a headline than real analysis.
Good campaigns, on the other hand, spend what it takes to get a good poll. Bad research means bad campaign strategy.
Polls are like anything else you buy. Some are good, and some are bad. Any number of things can cause a wrong poll: poor turnout sampling, poor-quality callers and just overall sloppy work
So, if I was editor of the N&O, I’d fire Research 2000 and hire another pollster.
Of course, this all may just be sour grapes: I relied on the Old Reliable’s poll. If I hadn’t, I’d have gone 9-1 on my election predictions instead of 8-2.
My other miss: I said Roger Koopman would beat Paul Coble for county commissioner. But Paul and Tom Fetzer did a good job of painting Paul as a pro-education Democrat. Let’s see if he votes that way on the board.
Also, a salute to Brad Crone and Ballard Everett, the two local consultants who ran the school bond campaign.
To comment, send us an email to comment@talkingaboutpolitics.com.
The N&O’s Poll Hole
There is a theory about that The News & Observer/WRAL pre-election poll on the
The explanation by the pollster – Del Ali of Research 2000 – was that the poll was accurate when it was taken two weeks before the election. But voters’ opinions changed.
I don’t buy it.
The same outfit had a poll several months ago showing the same result.
Plus, my experience is that voters who say they’ll vote against a bond issue don’t change their minds and then vote for it.
Two local polls – by Public Policy Polling and by the Home Builders of
There’s one question I’d ask the N&O about its poll: How much did you spend on it?
Too often, media outlets buy polls on the cheap. And they get what they pay for. They’re more interested in a headline than real analysis.
Good campaigns, on the other hand, spend what it takes to get a good poll. Bad research means bad campaign strategy.
Polls are like anything else you buy. Some are good, and some are bad. Any number of things can cause a wrong poll: poor turnout sampling, poor-quality callers and just overall sloppy work
So, if I was editor of the N&O, I’d fire Research 2000 and hire another pollster.
Of course, this all may just be sour grapes: I relied on the Old Reliable’s poll. If I hadn’t, I’d have gone 9-1 on my election predictions instead of 8-2.
My other miss: I said Roger Koopman would beat Paul Coble for county commissioner. But Paul and Tom Fetzer did a good job of painting Paul as a pro-education Democrat. Let’s see if he votes that way on the board.
Also, a salute to Brad Crone and Ballard Everett, the two local consultants who ran the school bond campaign.
To comment, send us an email to comment@talkingaboutpolitics.com.