Bookmark and ShareShare
Monday, March 15, 2010

Dueling Pollsters

The White House is pushing for a House vote on healthcare reform law by the end of the week. The New York Times predicts a week of “arm-twisting and high drama.”
And, we have a continuation of dueling claims about polls on this issue.

Two former Democrat pollsters penned an op-ed piece in The Washington Post. They claim that the battle for “public opinion has been lost.” The White House’s pollster shot back on Saturday in The Washington Post -- saying that “the American public is closely divided” on healthcare. [Note:  both pieces go on to talk about a number of other aspects of polling on healthcare as well.]

Here’s part of what the two Democratic pollsters (Patrick H. Caddell and Douglas E. Schoen) said:

"First, the battle for public opinion has been lost……. Nothing has been more disconcerting than to watch Democratic politicians and their media supporters deceive themselves into believing that the public favors the Democrats' current health-care plan. Yes, most Americans believe, as we do, that real health-care reform is needed. And yes, certain proposals in the plan are supported by the public. However, a solid majority of Americans opposes the massive health-reform plan."

And here's part of what the White House pollster, Joel Benenson, came back with:

"No pollster, including me, could look at the recent data and responsibly say anything other than that the American public is closely divided when it comes to supporting or opposing various health-care plans."

So we have "solid majority of Americans oppose" on the one hand. And "American public is divided" on the other.

On this point, I give the nod to Benenson. The recent available data do not show that a solid majority of Americans oppose the massive health reform plan. "Divided" comes closer to the mark.

Gallup’s latest survey was conducted March 4-7. We found that 48% of Americans would advise their representative in Congress to vote against "a healthcare bill similar to that proposed by President Obama"; 45% would vote in favor. The other most recent poll, conducted by AP-GfK at roughly the same time, found 43% would oppose and 41% would support “the health care reform plans being discussed in Congress.” (This poll offered respondents an explicit “neither support nor oppose” alternative, taken by 12%, with another 4% saying they were unsure.)

How to characterize these results? I might say “tilt slightly against” as a little more precise summary than "divided." But, in any case, I would not agree with the Caddell/Schoen interpretation that a “solid majority” of Americans opposes the legislation. At least, not based on the most recent poll data.


A side note. All of this highlights one of Benenson's challenges. His employer, the White House, is attempting to proactively pass a new bill. But Benenson can't use poll data to say that a majority of Americans want the healthcare bill passed. So he is, in essence, reduced to arguing that the other side is wrong if they say that the majority oppose passing a healthcare bill. In this case, to say the public is "divided" is a positive thing.
 
Clearly a lot of this is a glass half empty/half full situation. Caddell and Schoen are correct that the battle for public opinion on healthcare has not been won. Benenson may be right that it has not been totally lost either. The data still suggest that if the bill were put to a national referendum with a full turnout, it would lose.

Benenson has been active on the memo front. Before publishing his Saturday Washington Post piece, Benenson wrote and circulated a memo on public opinion and healthcare that was reportedly being widely read on Capitol Hill. (The Washington Post obtained and posted a copy of the memo.)

Benenson’s memo appears to have been designed to make Democratic representatives in Congress feel better about voting for President Obama’s healthcare bill. Benenson wrote a similar memo in September.

These are political memos. Not to be mistaken for scholarly meta analyses of all available evidence.

Benenson’s initial point in his memo:

“Recent public polling shows a trend of support starting to grow for the health care plan now under consideration while opposition to reform is on the decline. In fact, most recent public polls show support and opposition to the plan is roughly even.”

Two points here. Recent polls show “roughly even” support. And that the trend is toward support.

As noted above, the first point (which Benenson makes his second point) is generally ok -- if one sticks to the very most recent polls.

Strangely (at least, strangely to me) Benenson doesn’t, however, stick just to the most recent polls. Instead, he reaches back and lists two older polls to bolster his “roughly even” point.
To be specific, Benson lists four polls, as follows:

Support/Oppose President’s Reform Plan in Selected Public Polls


Poll Dates Support/Oppose Margin

AP/GfK 3/3-3/8 41/43 -2

Gallup 3/4-3/7 45/48 -3

Kaiser 2/11-2/16 43/43 Even

ABC/WashPo 2/4-2/8 46/49 -3


The first two, as noted, are the most recent available as of this writing. So far so good. These do show a small gap between those opposed and those in favor.

But the other two polls? Not clear why Benenson puts these two month-old polls into the mix.

There is, in fact, a fairly large corpus of healthcare reform legislation polls conducted in February. (Go here and here to see lists of these polls). The two February polls Benenson selects just happen to be ones showing a close margin. There are a number of other February polls which Benenson does not reference. He does not quote a CNN poll from about that same time showing that 73% of Americans -- given a three-part choice -- think that Congress should stop working on its current bill and work on an entirely new bill or stop working on a bill altogether. He does not quote a Pew Research poll from about that same time showing a 12 percentage point gap against passage of a bill. He does not reference a Newsweek poll from mid-February showing a 9-point gap against.

3 comments:

Anonymous said...
March 28, 2010 at 7:28 PM  

Recently, we are seeing results of polls showing that about half of Americans support the Obama health plan and about half are opposed. In order to clarify this result so that it can be interpreted much more meaningfully, can you please add a qualifier to the poll question. In particular, instead of just asking if an individual is for or against the new health plan, why not find out first if the individual pays federal income taxes? Obviously, if someone is not now paying taxes, then what difference does it make to them if the government increases spending/taxes. It is a critical piece of information to find out how those who actually pay the bill feel about the health bill. Please modify your polls accordingly to obtain this critical distinction among those who are polled.

Anonymous said...
March 30, 2010 at 1:09 PM  

Again where did these eroneous poll results come from.

FACT: Over 85% of Americans are AGAINST the so-Called health care reform!

This not guessing nor pretense, just that the majority IS against it!

Anonymous said...
March 30, 2010 at 1:50 PM  

What is evident in Benenson's blog, is he is out of touch with reality!

Benenson’s initial point CORRECTED by "JUST the FACTs PLEASE" sake:

“Recent public polling shows a trend of even greater opposition to the reform for the health care plan now under consideration while opposition to reform is exponentially on the increase as proved by the latter votes taking place to replace Yea sayers with Nay sayer- representitives against the reform "the public vote". In fact, most recent public polls show support and -opposition to the plan is roughly growing exponentially-.” i.e Massachusetts a Democratic state replacing their Demo with Republicans to STOP the reform!

Post a Comment

Comments are moderated by Gallup and may not appear on this blog until they have been reviewed and deemed appropriate for posting.

Copyright © 2010 Gallup, Inc. All rights reserved. | Terms of Use | Privacy Statement