Polling

As many people know, Rosie O’Donnell has landed herself in another high-profile brouhaha with yet another high-profile media figure.

In December, it was Donald Trump. Now, it’s Bill O’Reilly.

Yet, this time, her statements which precipitated the skirmish – that America may have set up the recent showdown between Iran and the captured British sailors, as well as the Bush administration being involved in bringing down the World Trade Center on 9/11 – appear to have enflamed the American people.

In fact, AOL is currently running an online poll asking the question (h/t Hot Air): "Should ABC fire Rosie, as O'Reilly implies?" The tally as of 10:00 AM EST Tuesday is:

At the end of his 2004 campaign, Howard Dean started Democracy for America, billed as a grassroots Democratic political action committee "dedicated to supporting fiscally responsible, socially progressive candidates at all levels of government."

Last week, DFA staged an online Dem presidential candidate preference poll. As a proud DFA member [meaning that I joined the email list], I just received the results of the poll. Hillary's advisers might want to dust off Hirohito's famous statement at the end of WWII to the effect that the situation "has not progressed entirely as we would have wished." Because Hillary finished in fifth place with 8.7%, behind Dennis Kucinich and barely one point ahead of Bill Richardson. Obama finished first with 28%, John Edwards a close second at 25%.

Now it's true that this is an unscientific poll, and that DFA surely attracts people from the liberal wing of the Dem party. But then again, isn't the conventional wisdom that Dem primary voters come from that same liberal wing?

Was DFA Executive Director Tom Hughes mortified by the results? He buried the mention of Hillary's embarrassingly bad performance in the eighth paragraph of his email!

There were two Iraq polls released on Sunday. One is guaranteed to be headline news. The other will likely be totally ignored.

In fact, one of the polls was already referenced by George Stephanopoulos on ABC’s “This Week,” as well as reported by USA Today and CNN.

Know what the difference is between these surveys, both of which rather compelling as they asked questions of Iraqi citizens? Well, one painted a rather dire picture of conditions in the embattled country, while the other found a very optimistic people who don’t believe their nation is in a civil war.

As the American media will likely focus all of its attention on the more pessimistic survey, here is the contrary view nobody other than Fox News is likely to cover as reported by the Sunday Times (emphasis added throughout):

Unbelievably, disgraced newsreader, Dan Rather, claimed at a recent festival that American journalism "has in some ways lost its guts" and that the MSM has "adopted the go-along-to-get-along (attitude)."

As reported by CNETNews.com, Rather was a keynote speaker at the South by Southwest Interactive festival this past weekend where he gave a 2 hour talk on the shape of journalism and the Internet.

One has to wonder to which "gutless" American media he is referring? Is it the same media that was so weak-kneed as to leak damaging national security information, the same media that just "goes along" to undermine the war effort at every opportunity? Is it the same one that goes out of its way to malign the US and Israeli governments? It is that MSM Rather imagines has somehow gone soft?

Brent Bozell's culture column this week unfolds the new polling numbers for the MRC's Culture and Media Institute on the American people's impression of moral decline and the media's role in it:

A new cultural-values survey of 2,000 American adults performed by the polling firm of Fabrizio, McLaughlin & Associates for the Culture and Media Institute reveals a strong majority, 74 percent, believes moral values in America are weaker than they were 20 years ago. Almost half, 48 percent, agree that values are much weaker than they were 20 years ago.

Push Poll: Definition: "A push poll is a political campaign technique in which an individual or organization attempts to influence or alter the view of respondents under the guise of conducting a poll. Push polls are generally viewed as a form of negative campaigning. The term is also sometimes used incorrectly to refer to legitimate polls which test political messages, some of which may be negative. Push polling has been condemned by the American Association of Political Consultants."

Melissa Russo, political reporter for NBC's NYC affiliate WNBC, recently followed Rudy Giuliani up to New Hampshire. In her report on this morning's "Today," Russo stressed that at a GOP campaign stop, Giuliani failed to inform the Granite State Republicans that "he's far from a social conservative."

View video here.

Despite their best efforts to woo the public, Democrats are still having trouble when it comes to convincing the public that they have a plan for the country. In the most recent Los Angeles Times poll, only 25 percent of the public believes the new Democratic majority actually has a plan.

This is one of several bits of bad (and good) news for Democrats which more than likely you will not see reported on the left-leaning TV networks. Read the rest of the poll, scroll to the bottom in this page, for details.

Free Republic, a conservative website familiar to many at NewsBusters, is normally not in the business of breaking news stories. FR's raison d'etre is to serve as a gathering place for conservatives to post and comment on stories found elsewhere in the media - including at NB.

But today comes some interesting investigative reporting at Free Republic, billed as an FR Exclusive. As reported by FReeper Howlin, CNN has recently signed a deal to provide poll data for the '08 prez contest with a polling firm controlled by an avid Clinton supporter.

Free Republic article here.

On Monday's CBS Evening News, White House reporter Jim Axelrod adopted Democratic terminology as he referred to President Bush's plan to call “for an escalation in the number of U.S. troops in Iraq.” Axelrod later applied the more Bush-friendly term of “surge,” but he also presumed the election was all about Iraq and not congressional Republican malfeasance in the Foley matter: “He'll be going against what voters said they wanted last November, and what the new Democratic leaders in Congress say they want right now.”

Axelrod soon highlighted how “a new CBS News poll out tonight indicates nearly six in ten Americans either want troop levels lowered or a full withdrawal." An on-screen graphic pegged the numbers at 59 percent wanting troops in Iraq “decreased or removed” and 35 percent wishing them “increased or the same.” But in citing only that one poll finding which combined answers, Axelrod skipped how the public is nearly evenly split on a “short term troop increase to stabilize Iraq,” with 45 percent in favor and 48 percent opposed. Furthermore, the PDF of the poll results reported that “those with friends and relatives who have served in Iraq are a bit more supportive of the idea than Americans overall” with 50 percent in favor and 44 percent opposed.

For creating a story out of nothing and then finger pointing at US society and saying how evil it is, this Dec. 12th CNN story takes the cake. In "Poll: Most Americans see lingering racism -- in others", not only is a somewhat leading poll cited as evidence that America is still rife with racism, but CNN uses comments emailed to them by their viewers as some sort of follow up proof for it!

Very scientific, I know. After all, CNN used science via the Internet and phone lines to conduct this farcical poll, I suppose.

(CNN) -- Most Americans, white and black, see racism as a lingering problem in the United States, and many say they know people who are racist, according to a new poll.

But few Americans of either race -- about one out of eight -- consider themselves racist.

The ink is barely dry on the Iraq Study Group's report and the mainstream media has already decided that the "bipartisan consensus" is gospel.

Just in case any neanderthals who might question this august group haven't gotten the message, MSNBC has created a stunningly blatant online push poll to get them in line.

In one of those "analysis" pieces reporters love to write, Washington Post White House reporter Peter Baker underlined on Wednesday one reason why NBC might have started using "civil war" to define Iraq: it severely undercuts the Iraq war in opinion polls.

Sen. Jack Reed (D-R.I.), a member of the Armed Services Committee, said Bush would rather frame it in the terrorism context to preserve public support. "If it's a civil war and only a small portion of it involves al-Qaeda operatives, then it's suddenly not the central front in the war on terror, it's a struggle by Iraqis for political power," he said. "That means the rationales for this are severely undercut."

Polls suggest that most Americans have already settled this debate in their minds -- 61 percent of those surveyed in September by NBC News and the Wall Street Journal described the situation in Iraq as a civil war, while 65 percent agreed in a CNN poll and 72 percent in a Gallup poll. Of those who described the conflict as "out of control" and a "civil war" in a later Gallup-USA Today poll, 84 percent called U.S. involvement a mistake, compared with 25 percent of those who did not view the situation that way.

What a difference a week makes – along with a huge gaffe by a former presidential candidate, and some unexpected changes in polling results.

The New York Times has published an Election Day article entitled “For Democrats, Even a Gain May Feel Like a Failure” (hat tip to AJ Strata) that dramatically lowered the bar of expectations for the Party of Pelosi and Reid. The piece by Adam Nagourney ominously began:

In most midterm elections, an out-of-power party picking up, say, 14 seats in the House and five seats in the Senate could call it a pretty good night.

But for Democrats in 2006, that showing would mean coming up one seat shy of taking control of both the Senate and the House. And it would probably be branded a loss — in the case of the House, a big one.

Wow, Adam. Tell us what you really think:

Of late, we've been treated to the sound of careless political reporters and analysts tossing around figures of Democrats taking back the House by varying degrees of hugeness.

In many cases, these predictions are simply wishful thinking on the part of left-wingers eager to see Republican evil (the only kind there is) beat back by the forces of good. Some of our prognosticating pundits are basing their predictions on actual polling, however.

Trouble is, their math, based somewhat on faulty "generic ballot" polls and primarily on the work of political analysts Stuart Rothenberg and Charlie Cook, is downright horrible.

How accurate are polls at predicting a winner? Not too. So long as a candidate is within 10 points, most polls shouldn't be readily relied on as predictors for who will win. Charles Franklin, a political science professor at the University of Wisconsin has an interesting post today about just how important the "margin of error" really is.

On a graph, Franklin compares poll results with actual election results, resulting in several observations, one of which is the importance of realizing that polls cannot reliably predict races that are less than 10 points apart.

One interesting feature is that a margin of zero (a tied poll) produces a 50-50 split in wins with remarkable accuracy. There is nothing I did statistically to force the black trend line to go through the "crosshairs" at the (0, .5) point in the graph, but it comes awfully close. So a tied poll really does predict a coin-flip outcome.

The probability of a win rises or falls rapidly as the polls move away from a margin of zero. By the time we see a 10 point lead in the poll for the Dem, about 90% of the Dems win. When we see a 10 point margin for the Rep, about 90% of Reps win. That symmetry is also not something I forced with the statistics-- it represents the simple and symmetric pattern in the data.

More practically, it means that polls rarely miss the winner with a 10 point lead, but they DO miss it 10% of the time.

A 5 point lead, on the other hand, turns out to be right only about 60-65% of the time. So bet on a candidate with a 5 point lead, but don't give odds. And for 1 or 2 point leads (as in some of our closer races tomorrow) the polls are only barely better than 50% right in picking the winner. That should be a sobering thought to those enthused by a narrow lead in the polls. Quite a few of those "leaders" will lose. Of course, an equal proportion of those trailing in the polls will win.

So read the polls-- they are a lot better than nothing. But don't take that 2 point lead to the bank. That is a failure to appreciate the practical consequences of the margin for error.

Sunday's lead story from Adam Nagourney and Robin Toner casts Republican party prospects in unrelieved grim tones ("G.O.P. Glum as It Struggles to Hold Congress -- Lost Seats Seen, Even in Best Outcome").

"The battle for Congress rolled into a climactic final weekend with Republican Party leaders saying the best outcome they could foresee was losing 12 seats in the House. But they were increasingly steeling themselves for the loss of at least 15 seats and therefore control of the House for the first time in 12 years.

As Ace’s “CNN To The Rescue” headline both accurately and comically depicted, the most trusted name in news released a poll of its own on Monday with significantly different numbers than those released by three other media outlets yesterday as reported by Newsbusters here, here, and here. Not surprisingly, much like results shared on Saturday by Newsweek, CNN’s polling organization, Opinion Research Corporation, found that support for Democrats in the upcoming midterm elections has expanded in the past week by nine percentage points.

Yes, folks, you don’t need to wash your monitor's screen. CNN said there's been an increase in Democrat support of nine percentage points in the past week. Don't believe it? Well, read it for yourself:

This is pretty hysterical. On Sunday, three different polls – Pew Research Center, USA Today/Gallup, and Washington Post/ABC News – were released showing evidence that the preference for Republicans in the upcoming midterm elections has been surging in the past couple of weeks, with the gap favoring Democrats now down to between four and seven percentage points. However, Newsweek released the results of its own poll Saturday, and the magazine actually sees the Democrats expanding their lead. I kid you not (emphasis mine throughout):

Imagine if you will that a month ago, a major newspaper, in combination with a major polling organization, had pronounced that the Republicans were ahead by 23 percentage points in voter preference for the upcoming midterm elections. Further imagine that just two weeks ago, this lead had been trimmed to thirteen. And, just for argument’s sake, with two days to go before the pivotal elections, the Democrats had cut this lead to only seven points. Do you think the opening paragraph in the article on this subject by this major, left-leaning newspaper might address this?

Well, USA Today just posted an article at its website (hat tip to The Strata-Sphere) concerning a new poll done with the Gallup Organization (this author is waiting with baited breath for the full results to be published!), and the most important finding of the survey didn’t come until the sixth paragraph (emphasis mine): “What's more, President Bush's last-ditch push for votes and Sen. John Kerry's comments that seemed to denigrate the education level of U.S. forces in Iraq have helped energize GOP voters. A Democratic advantage of 23 percentage points a month ago and 13 points two weeks ago is now down to 7.”

Instead of leading with that important information, the reader had to wade through the following:

Another poll was released on Sunday showing a late surge of support for Republicans in the upcoming midterm elections on Tuesday (grateful hat tip to NB reader American Infidel). This time it’s the Pew Research Center defying conventional wisdom (emphasis mine throughout):

A nationwide Pew Research Center survey finds voting intentions shifting in the direction of Republican congressional candidates in the final days of the 2006 midterm campaign. The new survey finds a growing percentage of likely voters saying they will vote for GOP candidates. However, the Democrats still hold a 48% to 40% lead among registered voters, and a modest lead of 47%-43% among likely voters.

The summary article continued:

In my first piece about this surprising Washington Post/ABC News poll published on Sunday indicating that the Republicans have been picking up ground on the Democrats in the past two weeks, I said that it would be interesting to see how this survey got reported. As compared to what ABC’s “This Week” did Sunday morning (i.e. beginning the program discussing it), CBS’s response was much more predictable. However, what was peculiar is the person CBS used to discredit the data given his pedigree and bona fides. 

With that in mind, Bob Schieffer invited CBS political analyst Stuart Rothenberg on Sunday’s “Face the Nation.” Rothenberg made it clear that he sees a big Democrat victory in the House on Tuesday (up to 40 seats), and the Democrats picking up four to seven seats in the Senate (video here). As the discussion moved to who will actually turn out to vote, Rothenberg questioned the methodology of the Washington Post/ABC News poll:

With just a little over a week to go NBC's Today show is cranking up the anti-Bush sentiment and this morning Meredith Vieira and Chris Matthews hammered how Republicans are running away from the President. After NBC's Kelly O'Donnell aired a full report on how Bush is showing up in ads everywhere for Democrats but not for Republicans, Vieira asked MSNBC's Hardball host Matthews to break down the midterms which just allowed Matthews to go on his typical anti-Iraq war tear:

Meredith Vieira: "I know you heard Kelly O'Donnell's report, the President now being featured in more than 160 ads for Democratic candidates. Do you think the strategy is gonna work?"

In his syndicated column, political demographer Michael Barone looks at some of the pratfalls of polling. He also makes some important points about how most media polls are overcounting Democrats:

In 2004, the electorate that went to the polls or voted absentee was, according to the adjusted NEP exit poll, 37 percent Democratic and 37 percent Republican. In party identification, it was the most Republican electorate since George Gallup conducted his first random sample poll in October 1935.

But most recent national polls show Democrats with an advantage in party identification in the vicinity of 5 percent to 12 percent. Party identification usually changes slowly. Historically, voters have switched from candidates of one party to candidates of the other more readily than they have changed their party identification.

I was a little stunned to hear one of leftist radio host Stephanie Miller's crew today hype that a new Newsweek poll shows 51 percent of Americans now favor impeaching George W. Bush. What? Perhaps they "learned" from the Daily Kos, which today tried this game of statistical wizardry: if 28 percent say impeaching Bush is a "top priority," 23 percent say it should be a lower priority, a stunning majority now "supports" impeachment. But the poll did not ask impeachment, yea or nay. Would an actual up-front question draw a lower number, since 23 percent chose an option that suggests "not so hot on that idea"?

Newsweek actually reported:

It seems the closer we get to Election Day, the more polls we’re going to be subjected to that over-sample Democrats. The latest entrant comes from Newsweek (hat tip to Stop The ACLU). In this one, 24 percent more Democrats were surveyed than Republicans.

Frankly, I have no interest in sharing any of the results from this survey, for if a polling agency can’t create a sample that accurately reflects the electorate, why should anybody care about the data it produces?

Maybe more importantly, why would a news organization pay for such results? After all, there are now 300 million people in America. It shouldn’t be hard to find equal numbers of Republicans and Democrats to answer questions, especially when the total sample is only 1,000.

In this instance, as reported by Reuters:

If you've ever wondered who it is that seems to be getting polled, or how come questions always seem to understate support for Republicans and conservatives, read this posting from Freeper "MHT" which tells of his encounter with a pollster:

Twenty-five years after registering to vote, I finally got polled by a national news group. And, as usual, Rush was right. However, there are a few things which I want to share.

1. The interviewer really wanted to talk to me. I was on another call and told them to call me back in 10 minutes. They did, indicating to me that they are having difficulty reaching people who wish to talk to them.

2. The questions were phrased in a very subtle way that focused on an anti-Bush agenda. For example, "Do you think that Bush is responsible for the situation in Iraq?" That's a yes/no question. They could have made it multiple choice. At best, Bush gets a 50% chance of being blamed for Iraq instead of insurgents, radical Islam, outside funding, etc.

These days, it is almost as telling what little gems media organizations choose to hide from the public about their own polls as what they share. The release of the most recent NBC/Wall Street Journal poll is a fine example. As NewsBusters reported, the good folks at the “Today” show on Thursday seemed quite giddy over numbers that suggest the Republicans are in a lot of trouble in the upcoming midterm elections. However, as is typical, what wasn’t shared from this study conceivably gives a different picture.

For example, as is typical these days, news organizations don’t like to share the political affiliations of those questioned. Certainly, you can’t blame them, for this might give the public some pause to trust the veracity of the data. This instance was no exception, for those that were either “strong Democrat,” “Not very strong Democrat,” or “Independent/lean Democrat” totaled 43 percent of the respondents. The tally for “Strong Republican,” “Not very strong Republican,” and “Independent/lean Republican” was only 37 percent. As such, 16 percent more Democrats or those who leaned Democrat were polled versus Republicans and those who leaned Republican. Color me not surprised.

But, that’s only the beginning.

What a difference a day makes!  Just yesterday, as I reported here, pollster John Zogby was boasting of his affiliation with Nova M, the new liberal radio network, heir to the moribund Air America.  The headline of the mass email he sent out proclaimed: "Air America Co-Founders Start New Liberal Talk Network: Mike Malloy, Joe Trippi and John Zogby join Nova M Radio."

As he breathlessly announced the "formation of [the] new progressive talk radio network" by a couple of Air America's original founders, Zogby touted his own connection, letting us know that he will be co-hosting a weekly one hour show, “The Pulse of the Nation.” The idea will be for Zogby to "poll particular hot button issues from politics and pop culture to the War in Iraq and stem cell research. Each program will include expert guests and audience participation. At the end of each show John will reveal the secret results of the poll and each listener will then know whether or not they have their finger on 'The Pulse of the Nation!'"

It is Wednesday, which means it was time for another installment of "Capitol Bob," Bob Schieffer; on CBS’ "Early Show." Today’s segment dealt with the midterm elections and Schieffer’s predictions. Schieffer still believes the Democrats will win control of the House, but cautioned Democrats not to get too over confident, as there are three weeks to go and the Republicans have a tremendous fund-raising edge. And in fact, the cash advantage was the only positive sign Schieffer could see for Republicans. Co-host Rene Syler, however, highlighted that with the Mark Foley coverage subsiding, experts were a little more skeptical of the Democrats chances. Syler introduced the segment:

"For weeks now, negative news from Washington left the Republican party sagging. But, now that it's gotten quieter, the GOP is ready to fight for control of the next Congress. And of course, for the inside scoop from Washington we turn to 'Capitol Bob,' CBS News Chief Washington correspondent and host of 'Face the Nation,' Bob Schieffer. Bob, good morning."

David Broder and Dan Balz wrote a rather lengthy, front-page story for the Washington Post this morning with the cautionary headline “Poll Shows Strong Shift Of Support to Democrats.” However, Broder and Balz chose not to share some key information from this poll with their readers, the most important of which being the political breakdown of those questioned. In fact, the meager percentage of Republican respondents to this survey should have led the Post to headline this article "Poll Shows Strong Shift Of Questions to Democrats!"

The article began: “Democrats have regained a commanding position going into the final weeks of the midterm-election campaigns, with support eroding for Republicans on Iraq, ethics and presidential leadership, according to a Washington Post-ABC News poll.”

Yet, nowhere did the authors let their readers know that 41 percent more Democrats were questioned for this survey than Republicans. That’s right. The breakdown was: