|
There has been a lot of posting about the late CBS and WaPost polls. I think both sides are misinterpreting (or perhaps misrepresenting for advantage) what polls do and are useful for.
First, polls prove nothing, except that those people surveyed responded in a particular way at a particular point in time to a particular set of question.
The polls do not prove Dean has the nomination. The polls do not prove Bush* et al will be re-elected. If I had a polling system that worked that way, I would point it at the horse racing track and not at the political system, and be the next George Soros.
What polls do tell you is how, over time, how those polled (and be extrapolation the population in general) have changed their views for an identical question set and sample selection, and perhaps in response to certain specific events external to the poll. But 1) you either have to make an educated guess about the external event (ex. Saddam's capture) or your have to change the poll, which is disruptive of the accuracy of your trend lines.
Also, polls can help one make a decision about message and audience between the time the poll is taken, and the target date (in our case, the election). In the case of the two most recently debated polls, what we see most of all is: Dean, as the likely nominee, has to figure how the hell to deal with the War in Iraq issue.
It doesn't tell him what to say, how to say it or what his audience is. But it does tell us that somehow, between now and next November, the Dean campaign and opponents of Bush in general, need to change the public perception of the Iraq War from a Great Victory for Democracy into something bad.
Clark, by coming out against the war forcefully, has the same problem. I think the other candidates less likely to win the nomination, but obviously Gephardt and Kerry have a different set of problems.
Events may do this for us, but should not be counted on. I trust the Rove Machine to make a very pretty Potemkim Democracy (or at least some sort of self-ruled Republic with some Democratic trappings) in time for the GOP Convention to Election Day window.
But I digress.
The polls are a useful tool in telling us 1) where we came from and where we're going and 2) how to get to where we want to be.
They are not oracles. They predict nothing, as prediction implies knowledge of the future, and polls are a snapshot of the past.
So before you hit that Post message button to trumpet your own candidates latest victory, or panic over what looks like a difficult task in winning the general election, I wish we would all spend more time talking about how any likely nominee overcomes the current situation, because the two leading contenders both need to do that.
If Gephardt or Kerry surprises us in the new year, then we will have a different set of issues and polls to talk about.
|
Printer-friendly format
Email this thread to a friend
Bookmark this thread
(1000+ posts)
Send PM |
Profile |
Ignore
