Jon Keeling’s post on today’s Quinnipiac poll is a lesson in bad propaganda.? It reminds me of how ever bad poll was spun by the McCain campaign to those infamous words:? “This is great news … for John McCain!”
Keeling tries his best to spin bad news, but he shouldn’t have bothered.? By simply trying (and failing) he reveals how Kasich has to resort to some major fallacies to put lipstick on this pig.
Keeling fallacy #1: “Comparing Quinnipiac’s ‘registered voter’ results to Rasmussen’s ‘likely’ voters results show that the more people know Kasich, the better he polls.”
He repeats this fallacy twice in his post.? He’s wrong.? Except to average the results, no serious pundit or polling analysts compares polls from different polling organizations like Keeling tries.? It’s an apple and oranges comparison.? Besides just a difference in the type of voter polled (registered vs. likely voter model), there are significant difference between methodology.? Quinny, I believe, uses live question takers, Rasmussen uses automated dialers asking recorded questions.? It’s long been held by pollsters that using automated dialers instead of live questioners lead to different results.? These differences make any attempt to compare one poll from the other and make assumptions from the differences statistically unreliable.
Regardless of the differences between the methodology of the poll, Rasmussen and Quinnipiac show that the more Ohioans become familiar with Kasich, the less likely they support him.? If you compared each polling outfits most recent poll to that same outfits prior poll of this race, you’d see that Kasich has made marginal improvement in his name recognition, but his standing has slumped, not increased.?
Since the most recent Quinny poll, Kasich has lost three points in Rasmussen against Strickland and five in Quinny.
The reason Kasich does better now with voters who are familiar with him that such voters are overwhelmingly white, conservative Republicans who are his base and the people he’s been exclusively courting for his campaign for the past THREE YEARS.? That’s hardly an accurate representation of the voting public in Ohio at large.
Keeling fallacy # 2: “It’s also clear that the massive Democrat [sic] efforts to define Kasich first have so far failed.”
Keeling repeats this theme throughout his post.? There are actually two fallacies (and one glaring example of ignorant grammar) with this Kasich talking point.? One, there has been no massive Democratic effort to define Kasich.? If Keeling thinks my posting about Kasich’s reckless tax plan, and ODP’s Kasich Tax Calculator is a “massive effort,” then the Kasich campaign is in for a shock when the Strickland campaign, ODP, unions, and other go from relying solely on free media to frame Kasich to a paid media model.? When we start seeing wall-to-wall ads attacking Kasich’s tax plan, then we can talk about “massive efforts” to define Kasich.? We aren’t there yet.
To date, the effort to define Kasich can be attributed mostly to this blog since December.? And even I, in all my vanity, do not believe that my writing of Kasich constitutes a “massive” Democratic effort, or the three to five point bounce Strickland got.
The second fallacy is that it, or something else, obviously has worked.
Since December Kasich has lost a THIRD of his lead in Rasmussen.? According to the most consistent polling organizations in this race, Kasich has lost anywhere between three to five points in a head to head matchup against Strickland since December– when I began my campaign to get the media to incorporate the fiscal implications of Kasich’s tax plan into the narrative of the race.
By calling it “massive,” Keeling implicitly concedes that Kasich and Keeling have been outflanked on message and that the Kasich campaign has spent the last three months on defense, not offense.? When the best he can say is deny that it worked (even though the polls show movement), you know he’s been worried that we, not his candidate, have been dictating the narrative.
Keeling fallacy #3:? No mention of the lack of a Mary Taylor factor.
Mary Taylor was picked for two political reasons: 1) get the GOP behind Kasich, and 2) make the Kasich ticket attractive to female voters.
Umm, faceplant.? Quinnipiac notes that Strickland’s improvement comes largely from his strength with female voters.
Also, according to Quinny, Kasich has more tepid support with self-identified Republican voters than Strickland does with Democratic voters.? In fact, Kasich loses Republican voters to Strickland already at half the level Ken Blackwell lost them in 2006.
In short, the poll shows that Democrats are more united behind Strickland than Republicans are against the GOP nominee.
Furthermore, I wouldn’t be surprised to find out that a large reason Kasich doesn’t have as strong of party unity as Ted Strickland is due to the utter chaos putting Mary Taylor on his ticket caused for the rest of the GOP statewide ticket.
Keeling fallacy #4: “On every single question, dissatisfaction and disapproval of Governor Strickland worsened from the last Quinnipiac poll last November.”
Is that true, Assistant Director of the Quinnipiac Polling Institute?
?There has been an improvement in voters? views of Gov. Ted Strickland,? said Peter A. Brown, assistant director of the Quinnipiac University Polling Institute. ?The movement is a few points, but it is consistent across a number of measures.”
In fact, Strickland has IMPROVED on the very question of whether voters approve of Strickland.? His disapproval and unfavorability ratings?? Yeah, on Strickland they went DOWN.? Keeling has the data completely backwards.
You know who’s unfavorability ratings did go up, though?? Yeah, John Kasich’s.
Keeling fallacy #5: “Strickland and the ODP have failed at improving Strickland’s reputation as a manager of Ohio’s economy and its budget.”
Um, again, Keeling cherry picks the data.? On the more important question of which candidate do voters believe would be best at handling the economy and the budget, Kasich (who voters admittedly don’t know) comes out ahead by six points.? However, his numbers are exactly unchanged from the last time Quinny asked that question while Strickland’s numbers have improved by two points on each question.
Also, another 2% volunteered that neither candidate would do particularly better on these issues, essentially reducing Kasich’s sole issue advantage to a near statistical tie.
Keeling fallacy # 6: “Ultimately, there is little evidence that the one thing that can help Strickland’s chances, a substantively improved economy, will happen over the next few months.”
The recession is officially over.? Fourth quarter GDP ’09 was estimated at 5.7% increase.? The 3Q showed a 2.2 increase. Housing prices are going back up.? Housing starts rose 21.2% last year, the largest year-over year increase since April 2004.? Retail is beating expectations.? Unemployment is a lagging indicator.? Keeling and the rest of his “Cheerleaders for continued economic failure” who have made a political bet that the economy wouldn’t improve substantially late last year are losing big.
Keeling can try to spin this news all he wants, but it’s clearly nobody is buying that this isn’t evidence good news for Strickland:
- Politico: “Strickland retakes lead over Kasich”
- National Journal: ” Strickland Reclaims Lead, Still Under 50″
- CNN: Strickland’s numbers on the rebound”
- MSNBC “First Read“: “Buckeye State Watch: Is the worm beginning to turn for Democrats in Ohio, too?”
Oh, and you know all those conservative bloggers who have been trying to make this race seem like last year’s gubernatorial race in New Jersey?
“It?s one poll, but it might give pause to Republicans who think the 2009 off-year elections and 2010 Massachusetts race point to an electoral tsunami that they can ride to easy wins. Strickland is in far better shape than, for example, former Gov. Jon Corzine (D-N.J.) was at this point in 2009 ? last February, he trailed now-Gov. Chris Christie (R-N.J.) by six points.”–David Weigel (formerly of The American Spectator, The American Conservative, &?The American Prospect), Washington Independent?(“In Ohio, a Republican Candidate Slips in the Poll.”)
And what polling outfit showed Corzine losing by six points at this point a year ago?? Quinnipiac.