Pollster.com

June 16, 2007

Re: Pardon?

By Mark Blumenthal

An update to my post yesterday on speculation about how a pardon of Scooter Libby might affect the Bush job rating: The amazingly responsive analysts at the Gallup organization sent this update, which answers the question I posed above by and shows how reactions to the notion of a Libby pardon back in March mesh with the Bush job rating at that time.

Among Republicans - March survey:

• 31% Approve Bush job / Favor pardon
• 35% Approve Bush job / Oppose pardon
• 9% Approve Bush job / DK-Ref on pardon
• 3% Not Approve Bush job / Favor pardon
• 16% Not Approve Bush job / Oppose pardon
• 5% Not Approve Bush job / DK-Ref on pardon
• 100% Total

As of March, only 3% of Republicans favored a Bush Libby pardon but did not approve of the job George Bush was doing as President. Meanwhile, roughly five twelve times as many (35% 16%) opposed a pardon but approved of Bush.

Looking at it another way, roughly nine out of ten of those who favored a pardon in March already approved of Bush's performance.

Of course, two big caveats are in order. First, as John Dickerson pointed out via email, a lot has changed in the Libby case since March. He has been sentenced to a term that especially angers his supporters and asked to serve his time immediately. We also now have three presidential candidates (Thompson, Romney and Giuliani) making the case for a pardon. So we might see different reactions among Republicans the next time pollsters ask the pardon question.

Second, commenter Chris G has a point when he argues that figuring out "the influence of a big media event like the pardon" requires "so much more than voters simply hearing a single sentence that's analogous with what polling interviewers [ask]." Using survey data to measure current attitudes or reactions is one thing. Using it to try to predict the future attitude is inherently speculative.

PS: Good catch twc -- sorry for the oversight

By Mark Blumenthal on June 16, 2007 11:32 AM | Permalink | Comments (3) | TrackBacks (0)

June 15, 2007

POLL: Texas Lyceum Statewide Primary/GE

By Eric Dienstfrey

A new newly released Texas Lyceum statewide survey of 1,002 adults in Texas (conducted 4/26 through 5/7) finds:

• Among 303 likely Republican primary voters, Sen. John McCain edges out former Mayor Rudy Giuliani (27% to 23%) in a statewide primary; former Sen. Fred Thompson trails at 11%, former Gov. Mitt Romney and former Speaker Newt Gingrich both at 6%.

• Among 362 likely Democratic primary voters, Sen. Hillary Clinton leads Sen. Barack Obama 33% to 21% in a statewide primary.

• General election match-ups:
  
  Clinton 35%, McCain 36%
  Clinton 31%, Giuliani 32%
  Obama 25%, McCain 32%
  Obama 22%, Giuliani 32%

By Eric Dienstfrey on June 15, 2007 8:25 PM | Permalink | Comments (1) | TrackBacks (0)

Pardon?

By Mark Blumenthal

In a column predicting a presidential pardon for former White House aide I. Lewis "Scooter" Libby, Slate's John Dickerson speculates about the potential impact on such a pardon on the job approval rating of President George Bush:

The number of people who would be angered by a pardon who haven't already abandoned the president could fit in an airport shuttle bus. Given the conservative defections from Bush over his support of immigration reform, a pardon of Libby-which would be popular with conservatives-might actually improve his approval ratings. Libby's conviction is seen as such an outrage among conservatives that one former Bush aide suggested "the consequences of not pardoning, if Scooter is led away in shackles, will be uglier than pardoning."

I'm not so sure. I have no doubt that Dickerson hears genuine outrage at Libby's conviction from the activist conservatives he talks to, but recent public opinion polls show little evidence of the same sentiment extending to the majority of rank-and-file Republicans.

Two survey organizations, Gallup and CNN/ORC, asked about a potential Libby pardon back in March:

Gallup (n=1,009 adults, March 11-14) - As you may know, a jury found Libby guilty on four out of five criminal counts. Do you think George W. Bush should or should not issue a presidential pardon for Lewis "Scooter" Libby?

21% Yes, should
67% No, should not
12% No opinion

CNN/Opinion Research Corporation (n=1,027 adults, March 9-11) - As you may know, Lewis "Scooter" Libby, Dick Cheney's former top aide, was recently found guilty of obstruction of justice and perjury in connection with the CIA leak investigation. Do you think George W. Bush should or should not give a presidential pardon to Libby?

18% Should pardon
69% Should not pardon
13% No opinion

While these results obviously show that, relatively few Americans reacted favorably to the notion of a Libby pardon in March, Dickerson argues that conservatives feel differently. So I asked for tabulations of those results by party, and here, thanks to a quick response from both Gallup and CNN, are the results:

So support for the Libby pardon is higher among Republicans. However, while roughly a third of Republicans supported a Libby pardon in March, more than half expressed opposition. Thus, "the number of people who would be angered by a pardon who haven't already abandoned the president" might fit more than a few airport shuttle buses.

The Gallup analysis does show slightly greater support for a pardon (44% support, 55% oppose) among the relatively few Americans (13%) who say they were following the issue "very closely" back in March. So perhaps we might find evidence of more "outrage" at the Libby verdict among those Republicans who have been following the story closely. However, I checked with Gallup, and only 14% of Republicans - and 4% of all adults -- fit into that category.

Perhaps support for Libby has been growing among Republicans since March? Unfortunately, neither Gallup nor CNN have asked this question since. However, the Rasmussen automated poll (which if anything may over-represent the opinions of well informed Americans) shows Libby's favorable rating at just 19% overall (and 34% among Republicans [sub. req.]) as of last week.

Of course, to really test Dickerson's theory, we would want to see how many Republicans are both unhappy with Bush and eager for a Libby pardon, and compare that to the number that currently approve of Bush's performance but oppose the idea of a Libby pardon. To do that, particularly in light of the continuing decline in the Bush ratings since March, we would need to work with more current results. But my guess is that even among Republicans, the net impact of a Libby pardon on Bush's job rating would not be positive.

One last point: Given the relatively low attention scores given this issue on Gallup's poll back in March (and even lower attention paid to the Libby trial as measured by the Pew Research Center in February), we can assume that the results above -- like many questions asked lately about immigration policy -- measure mostly "reactions" rather than currently held opinions. But in this case, reactions are what we care about. The question is how Americans would react to a Libby pardon, and while others may differ, I assume that coverage such a story would stress the same details regarding Libby's conviction as included in the questions quoted above.

UPDATE: Gallup sends the numbers that "really test" the theory plus more thoughts here.

By Mark Blumenthal on June 15, 2007 6:31 PM | Permalink | Comments (8) | TrackBacks (0)

June 14, 2007

Re [3]: Obama-Clinton "Virtually Tied" or Not?

By Mark Blumenthal

We talked a lot last week (especially here and here) about whether the most recent USA Today / Gallup poll, showing a surprisingly close result on the national Democratic primary trial-heat question, was measuring some real change or just the sort of random noise that comes with the territory of random sampling variation.

Whatever the reason, Gallup's Frank Newport dropped this big hint this afternoon on his Gallup Guru blog that last weeks' result will turn out to be something of an aberration: 

We have a Gallup poll now in the field, with results to be reported early next week. Preliminary indications are that this poll will find Clinton back in her typical leadership position as she has been for the most part this year.

If they report on their usual schedule, look for data to appear in the Tuesday edition of USA Today.

By Mark Blumenthal on June 14, 2007 10:15 PM | Permalink | Comments (1) | TrackBacks (0)

Who would you NOT support?

By Mark Blumenthal

In the comments section on my post about Hillary Clinton's support from women, reader timm0 asks:

Has anyone thought to ask this question in a poll:

I will read through the list of candidates. Please tell me which ones you will absolutely, positively NOT vote for?

I'd love to hear what the answers are for that.

Good question. Pollsters often ask exactly that question in large multi-candidate primaries. In fact, the NBC/Wall Street Journal survey asked it on their March 2007 survey (n=1,007 adults):

(ASK ONLY OF RESPONDENTS WHO SAY THEY ARE A DEMOCRAT IN Q.10a OR THAT THEY WOULD VOTE IN THE DEMOCRATIC PRIMARY IN Q.10b.)
Thinking of these same candidates again, are there any candidates for whom you would definitely NOT vote for the Democratic nomination for president?

16% Hillary Clinton
11% Dennis Kucinich
9% Joe Biden
7% John Edwards
7% Chris Dodd
7% Barack Obama
5% Wesley Clark
5% Bill Richardson
1% Other (VOL)
29% None (VOL)
17% Not sure

(ASK ONLY OF RESPONDENTS WHO SAY THEY ARE A REPUBLICAN IN Q.10a OR THAT THEY WOULD VOTE IN THE REPUBLICAN PRIMARY IN Q.10b.)
Thinking of these same candidates again, are there any candidates for whom you would definitely NOT vote for the Republican nomination for president?

20% John McCain
14% Newt Gingrich
9% Mitt Romney
6% Sam Brownback
7% Rudy Giuliani
4% Mike Huckabee
3% Tommy Thompson
2% Duncan Hunter
0% Other (VOL)
24% None (VOL)
22% Not sure

Remember, of course, that these results are a bit dated. The full results of the NBC/WSJ poll conducted this week have not yet been released, but those available as of this posting do not include an update on the "who would you NOT support" question.

By Mark Blumenthal on June 14, 2007 9:21 PM | Permalink | Comments (3) | TrackBacks (0)

Living in "Pollsterland"

By Mark Blumenthal

In her new "Poll Positions" column, CBS News polling directory Kathy Frankovic discusses the way pollsters love to ask questions that ask, "what if?" Although, as she points out, such questions were a favorite of polling pioneer George Gallup, they come with a potential danger:

[S]ometimes trying out those ideas may put us just outside of the real world and into what might be called Pollsterland.

It's not just that we can ask questions about Iraq, or health care, or immigration, or baseball's Barry Bonds, all of which are aspects of current reality. But we also can ask questions about an invented reality, such as: What would happen if . . .?

She goes on to list several examples, especially the "most famous" what-if of all, the one that begins, "if the election were held today..."

Frankovic also notes that Gallup started asking "as early as 1958...whether Americans would support an African-American candidate for president." My colleagues in the American Association for Public Opinion Research (AAPOR) picked up the discussion on our members-only "listserv," where someone pointed out that Gallup started asking 21 years earlier about whether Americans would support a woman for president.

This observation prompted one AAPOR member to recall the remarkably leading language used the very first time George Gallup asked a "what if" question about a potential woman presidential candidate. I checked the Gallup archives (sub. req.), and here it is along with the results (from a survey conducted in January 1937 among 2,888 adults nationwide):

Would you vote for a woman for President if she qualified in every other respect? [emphasis added]

33% Yes
64% No
3% No opinion

"Pollsterland," indeed.

PS: Thanks to AAPOR member SL for remember the 1937 Gallup question.

By Mark Blumenthal on June 14, 2007 9:00 PM | Permalink | Comments (1) | TrackBacks (0)

POLL: Harris Interactive National GOP Primary

By Eric Dienstfrey

A new Harris Interactive online survey of 3,304 adults (conducted 6/1 through 6/12) finds:

• Among likely Republican primary voters, former Mayor Rudy Giuliani leads former Sen. Fred Thompson (30% to 22%) in a national primary; Sen. John McCain trails at 18%, former Gov. Mitt Romney at 11%, and former Speaker Newt Gingrich at 8%.

• In a multiple-response question, 30% of all adults would consider voting for Giuliani for president, 21% for McCain, 19% for Thompson, 14% for Romney, 11% for Gingrich.