« the Internet & civil society | Main | some things a candidate could say ... »

August 30, 2004

reading polls

I don't pay too much attention to "point-estimates" in surveys (for example, Kerry is at 46% or Bush is at 47%). These results involve the usual margin of error, as in any random sample. To make matters worse, telephone surveys are becoming less reliable because many people have no land line or refuse to talk to pollsters. The unreliability is then literally multiplied because the point-estimate is a function of two questions, not one. Pollsters ask: "Are you a registered voter?" and then "Do you intend to vote for Bush or Kerry?" (They phrase both questions more carefully than this, of course.) Thus their bottom line is a crosstab based on two questions; and we know that the voter registration data are always quite inaccurate. Given these layers of bias, it's no surprise that even national polls conducted a few days before an election often fail to predict the popular vote.

While point-estimates are unreliable, trends in the same survey should be more meaningful. That's why I pay virtually no attention to anything except the Rasmussen Tracking Poll, which is the only public source of its kind. According to Rasmussen, the trend since August 1 is 2-3 points down for Kerry and 2-3 points up for Bush.

Why? Of course, no one knows. The only way to pursue this question seriously would be to find a random group of citizens who had changed their mind recently, and then ask them in-depth questions about why. In the absence of such information, we can only speculate. Some will claim that the shift is the fault of the Swift Vote Group--so Kerry should hit back hard. I think Ruy Teixeira has rebutted that theory. It's also unlikely that Bush has gained from the general news environment. On the contrary, the economic data, the situation in Iraq, and the fallout from Abu Ghraib have all been awful. Nor has the president said or done anything very impressive since August 1.

Rejecting those alternatives leads me to the theory that I want to believe anyway, for reasons of principle. I think the Kerry campaign has failed to look forward sufficiently. They have done an inadequate job of showing why the Bush policies for the next four years will be harmful, and--most importantly--they have failed to offer new policy ideas that are both plausible and inspiring. Tomorrow, I'll throw out some potential ideas.

Update: This is exactly the kind of backward-looking and negative message that I do not think Americans will swallow:

The Democrats do have a message but it's been submerged for most of the last three weeks. And that is the main reason why they've lost traction over that period.

The message is straightforward and explainable in ascending levels of specificity.

At its simplest: President Bush has screwed everything up.

August 30, 2004 10:40 AM | category: press criticism | Comments


Site Meter