After reading Tuesday's post on the new polling data on the Employee Free Choice Act (EFCA), I received the following comment from Guy Molyneux, the partner at the Democratic firm Hart Research Associates that directs their polling on behalf of the AFL-CIO:
I have to quibble with your excessively balanced statement on the EFCA polls: "While I am sure that both pollsters consider their questions 'fair and balanced,' partisans on both sides will see evidence of bias or leading language." Our question was a good faith effort to describe both sides' positions in the strongest language possible. Of course, that doesn't guarantee a perfect, or even the best possible question, but that's what we were trying to do. And given the low level of awareness you correctly describe, I do believe this kind of question is the best way to anticipate how the public will react to the debate, if/when they do tune in. Rasmussen's questions, on the other hand, look like they were commissioned by the Chamber of Commerce for public release.
To clarify: I did not mean to imply that the questions I reproduced from Hart/Molyneux and Rasmussen were intentionally biased or that either pollster was intentionally cooking the numbers. My point was that no matter how hard pollsters try to craft what they consider balanced questions, partisans resistant to their results will inevitably see bias in the language. This phenomenon is as true for journalism as for polling: The meaning of "balance" in characterizing two sides of a political debate is often in the eyes of the beholder.
That said, Molyneux raises a good point about the value of questions that present balanced summaries of the rhetoric on both sides of an issue. When we know for certain that the debate over an issue will ultimately attain a very high profile, those questions can do a better anticipating where public opinion is headed than more straightforward measures of current opinion. We can anticipate, for example, that lesser known but well funded candidates in highly competitive elections will inevitably become well known. So campaign pollsters find they get a better read on where the race is headed if they offer balanced descriptions of the candidates to see how preferences change.
The problem with most public policy issues debated in Congress is that they rarely get anywhere near the attention and awareness as a high profile election campaign. The immigration debate of 2007 was such an example. Even though it generated a lot of cable news coverage, only 18% of Americans told Gallup they were closely following the debate just a few days before it came to a vote in the Senate.
EFCA is likely to attain a higher profile than the 2007 immigration bill, mostly because two of the wealthiest and most powerful lobbies, organized labor and business, consider it a huge priority and are presumably poised to spend tens, perhaps hundreds of millions on paid advertising and outreach. But how much attention will Americans ultimately give this issue? For that answer, we will have to wait and see.