48% for Taco Bell, 52% for invading Iraq

Matt Taibbi, who writes about politics for Rolling Stone, among others, on what's wrong with polling:

In an interview with the San Francisco Chronicle on April 1, 2003, pollster Steven Kull said he believed that 40 percent of Americans were firmly behind the war, 20 percent firmly opposed it, and the remaining 40 percent supported it "either out of deference to the president or a sense of patriotism." He characterized the stance of the latter group as "pretty soft."

Well, no shit. Just as Kull predicted, the 40 percent firm-support number has remained an absolute constant since the beginning of the conflict. In the CNN/Gallup poll last week, that same 40 percent said they remained firmly in support of U.S. forces remaining in Iraq. Clearly, it's that "pretty soft" other 40 percent that's slipping. Those are the people I have a problem with, and it is with regard to those people that our polling system failed us two years ago and continues to fail today.

It seems fairly obvious that, in the course of the last few years, roughly 25-30 percent of the country has been influenced by the steady issue of news about increased violence and instability in Iraq. Apparently, a large percentage of Americans who supported the war two years ago have since become freaked out by the fact that, surprise, surprise, people are dying.

Which invites the question: If these people can't handle a few bad headlines, what exactly was their level of commitment to begin with? Pre-war polls, confined to the standard Coke-Pepsi either-or formula, didn't tell us much about that.

Maybe if the polls back then had been conducted differently, we might have had different results. Imagine a March 2003 poll that posed the following questions:

Would you yank your son out of college and send him to die for this bullshit?

Would you yourself be willing to give your life for this cause? If yes, grab your shit; there's a bus outside.

Read the whole thing on AlterNet.