Why Polls Make Americans Look Stupid

I'm always intrigued by blame-Americans-first analysis, so when I saw that Slate's Jake Weisberg had a new piece -- "Down With the People: Blame the childish, ignorant American public--not politicians--for our political and economic crisis" -- I had to click.

But I think the article reveals that he's after the wrong culprit for our political and economic malaise. Maybe it's neither the public, nor the pols. Maybe it's the polls.

Weisberg's key paragraph begins:

Anybody who says you can't have it both ways clearly hasn't been spending much time reading opinion polls lately. One year ago, 59 percent of the American public liked the stimulus plan, according to Gallup. A few months later, with the economy still deeply mired in recession, a majority of the same size said Obama was spending too much money on it...

There's a simpler reason Americans say they don't like the stimulus. Since the Recovery Act passed, the government spent $2 billion dollars and the unemployment rate went up almost two points. It's understandable -- if misguided -- to conclude that the Obama stimulus has been both expensive and unhelpful. But when pollsters ask us if we'd like to help the unemployed or improve our roads and bridges, well hey, that sounds like a pretty fine idea.

Sixty percent of Americans want stricter regulations of financial institutions. But nearly the same proportion says we're suffering from too much regulation on business. That kind of illogic--or, if you prefer, susceptibility to rhetorical manipulation--is what locks the status quo in place.

Weisberg is looking for hypocrisy in polls aren't comparable. The first poll finds that 60 percent of Americans wanted "new, stricter regulations on financial institutions" in November 2008. Of course they did: Lehman Brothers' bankruptcy nearly destroyed the financial system two months earlier. The second poll from 15 months later in January 2010 finds 57% of Americans are more concerned by too much regulation of business than too little. This poll came after many bank bailouts, auto company rescues, and a grueling health care debate later.

There are a couple of problems with proving the "Americans-want-it-all" thesis with polls. The first is that polls ask questions in different ways that can skew responses. The second is that many polls probe slivers of policy -- Would you like to reduce the deficit? (yes!) Would you like more tax money back to pay for your kid's education? (of course!) -- without forcing respondents to grapple with implications (ie Given that both tax increases and stimulus spending add to the already booming deficit, which of the following do you think would be a proper way to handle our long-term fiscal crisis?). Maybe Americans are know-nothing babies. I don't know enough people to know either way. But divining policy from individuals polls is like trying to paint a panoramic mural from two disparate Polaroids. It's just not a worthwhile experiment.