By Mike Dorf
I have an Op-Ed in the NY Daily News in which I argue that the people who have been lamenting the NY Times/CBS Poll about the Supreme Court's supposed unpopularity are missing the real story: The good news is that people have a realistic picture of the Court and they seem to accept that, on the whole, having judges subject to human emotions is useful. The piece expands on some of the themes I raised in a blog post last week.
Here I want to gripe a tiny bit about the standard polling question that asks respondents whether they "approve or disapprove" of the way some institution (e.g., Congress, the SCOTUS) or person (e.g., the President) is handling its or his job. It's remarkable to me that in answer to this kind of question many people answer yes. Maybe I'm just negative (though I don't think so as a matter of temperament) but if asked this question about Congress, the Supreme Court and the President my answers will be no, no and no. Of course, my reasons vary, but won't that be true of nearly every respondent. Wherever you are on the political spectrum, there will be some things that you disapprove about each actor.
I suspect, therefore, that the people who are answering yes are interpreting the question to mean something like the following: "All things considered, and in light of the politically realistic alternatives, do you approve of the job that the President [or Congress or the Supreme Court] is doing?"
But even then, I think the question is flawed, at least in principle. People who are unhappy with President Obama because they think he's a socialist will say they disapprove, and they will be lumped in with people who disapprove because they think that he has been too close to his immediate predecessor on national security.
The fact that there is nonetheless a substantial body of approvers (except for Congress) says to me that most respondents either do not have strong opinions about politics and/or do not follow the news closely enough to know whether the President has been pursuing policies that align with their political preferences. It's also possible that--in the case of the President's approval rating--the people (like yours truly), who think the President has been too centrist/conservative are such a tiny minority as to not show up in a poll that generally splits respondents on partisan grounds.
Accordingly, even if I were the czar of polling, I probably would not change the standard approve/disapprove question to something like the more nuanced hypothetical one I quoted above: Whatever small gains in fine-grained accuracy it would reveal would be wiped out by the loss of comparability with past polls.