Today's post on
polls for Santorum's seat in PA reminds me of the role early polls play in keeping the already influential in power, and setting up artificial barriers for newcomers. Instead of being about whether the newcomers can get their message out and persuade people, polls tempt us to assume that people's often-tentative opinion
before they've heard what any newcomer has to say, will be their opinion at election time.
Early polls are, for the most part, measures of name recognition, and vague past reputation. They can reasonably measure support for an incumbent people know, but they don't measure anything more useful about potential challengers. They just tempt us to think they do.
However, this does bring to mind one creative way I found to make use of early polls for the Democratic primary race, in 2003...
Early on, Lieberman was the candidate with the big name recognition. Other than Kerry, nobody in the field could even come close to challenging him. In all the polls, Lieberman was on top, and because of this, the media and the party were deluded into thinking he was a serious candidate.
I happened to know that he wasn't. I could tell he had no real support, and the only people saying "Lieberman" to pollsters were people who weren't paying any attention, hadn't given the primary contest any thought, and found his name familiar because he was Gore's VP choice. Right from the start, I predicted, and told people, that not only did Lieberman not have even the slightest chance of winning the nomination, he also didn't have a good shot at getting above 10% or winning delegates, in most states. (He ended up winning no delegates, and his only two-digit result was 11% in DE).
Knowing that made the early polls useful, not for measuring how the candidates were doing, but for measuring how the electorate was doing. This was the simple rule:
- Any poll showing Lieberman in first or second place, indicates that the population being polled, is not yet paying attention to the presidential primary.
- Once Lieberman settles into the single digits where he belongs, I knew the primaries had started in earnest in the state being polled.
Indeed, Lieberman plummeted to single digits in polls of NH and IA long before he did in national polls or those in any other state. The perceived reasons for this pushed by the news media made were just confused. The real reason was simple: Candidates were campaigning seriously in those two states, and the people being polled were starting to pay attention. So, very few of them were saying "Lieberman" to pollsters anymore. "Lieberman" was just code for "I have no idea who's running", for most people.
As the fall and winter of 2003-2004 rolled on, Lieberman began his plummet in state after state, roughly in proportion to their early primary dates and importance. And that was the indicator for when the rest of the numbers in the poll were worth paying any attention to.
Now, I don't know PA politics. I don't know enough to have strong, hard opinions about the levels of support for people like Santorum, Casey, Hoeffel, or whoever ends up in the final race. For people like me (and there are a lot of you out there), I think the best reaction to these sorts of early polls is to act as if you haven't seen them. Do not be tempted.
For those of you who do know PA politics well enough to have a strong confidence about the positions some of these candidates will be in when the summer and fall of 2006 come 'round, you might find good ways to use these polls, like I did above. If so, please tell us what you know. As long as what you know isn't based on these polls, because if it is, then I don't think you know it.