Push polls are an unfortunate reality of the political world -- shadowy creatures, shrouded in secrecy. Rarely do they see the light of day.
But this one has, and I offer it on dKos for education purposes.
The candidate is Lisa Quigly, who is running in the Democratic primary to replace her current boss -- outgoing Rep. Cal Dooley of Fresno (CA-20th CD). Her opponent is Jim Costa.
The race is otherwise unremarkable, especially considering it is safely Democratic. The only drama is in the Dem primary.
I somehow ended up on her email list, and got this today:
The poll of 500 likely Democratic voters conducted between Dec. 18-22, 2003, shows that once voters hear profiles of both candidates, Quigley gains 26 points to achieve 37 percent of the vote, while Costa drops down to 42 percent--a significant move of support away from Costa toward Quigley after only a positive description of the two candidates.
This is apparently an email to supporters and press. But here she makes the admission that she only gets 11 percent of support. Not too smooth. That's the kind of information a campaign keeps
really quiet. But it gets worse.
In addition, after voters hear a complete set of facts about each candidate's public record and history, Quigley continues to make significant progress and pulls ahead by five points, gaining a total of 30 points to win 41 percent of the vote compared to Costa's 36 percent and other 24 percent still undecided.
"Facts about each candidate" should immediately set off alarm bells. It's obvious those "facts" about Costa won't be too positive. But rather than keep us guessing ("was it really a push poll?"), she serves up the goods. In effect, the "push" questions, charging that Costa:
Voted to increase the car tax by over one billion dollars; (70% "less likely" to support)
Voted for California's failed energy deregulation scheme; (67% "less likely")
Did not support legislation to improve breast cancer treatment; (65% "less likely")
Was arrested for soliciting a prostitute; (65% "less likely")
Opposed bill to crack down on contractors who cheat
farmworkers out of wages; and (60% "less likely")
Switched positions mid campaign and dropped his
opposition to Prop 187 I don't have a horse in this race, and I don't know if the charges above are true or not. But it's amusing to me that Quigly has so openly admitted to push polling.
Given she's languishing at 11 percent of support, perhaps she's desperate. But push polling is not the sort of thing that anyone should brag about. Good thing she did, since it gives a rare glimpse into that shadowy world.
Update: I'm told by people in the know that I have flubbed on the definition of "push poll". As one person emailed me:
Your post about push-polling is way off the mark. Push polling is a phenomenon more often talked about than actually enagaged in. Push polls aren't real polls. Real polls use small, random samples of 500 or a 1000 people. Push polls are phone sweeps of large numbers of people, non-randomly selected, that push a message disguised as a poll question.
Fair enough. Still, call it "message testing" and say everyone does it (which apparently they do). It's still ugly, and if necessary, it's something best done quietly.
The beauty of this site is that everyone gets an education. Most of all me.
