Dialogues: Re: USA Africa Dialogue Series

In that parallel universe where Mitt Romney leads in all polls, Obama is also the devil, the anti-Christ, a Muslim, communist, Kenyan fifth columnist working to destroy the good ol' USA and fueled by Kenyan anti-colonial rage. Also, in that universe, Obama has buried America in a mountain of new taxes, legalized drugs and prostitution, has surrendered and apologized to America's enemies, and was only able to take out bin Laden because his fellow Muslims ratted out the terrorist mastermind to him in order to boost his reelection chances. Oh, did I also mention that in that conspiratorial world of crazy delusion, Obama is, in addition to being a Kenyan, also the son of a black American communist?

On Mon, Oct 1, 2012 at 8:50 AM, Toyin Falola <toyinfalola@austin.utexas.edu> wrote:

I doubt, but read a pro-Bangura analysis

The parallel universe where Mitt Romney leads all polls
By: Jonathan Martin and Alexander Burns
October 1, 2012 04:27 AM EDT

To talk with any working Republican political operative these days is to hear the same tale of woe: a grim accounting of the past few weeks, a dash of gallows humor and a measure of hope that President Obama is still beatable. Never in question is that Mitt Romney is trailing — the private surveys these strategists see for their down-ballot clients make that clear. The only question is by how much.

But hanging up the phone or clicking out of e-mail is to find a parallel universe on the right. On TV, talk radio and especially the Internet is a place where the swing-state polls that show Romney losing are not just inaccurate but part of an intentional plot by the heretofore unknown media-pollster axis to depress Republican voters. In this other world, Romney not only isn't losing — he's on the verge of a convincing victory.

"I believe if the election were held today, Romney would win by 4 or 5 points," trumpeted Dick Morris on Fox News last week, predicting a win for the GOP ticket in Florida, Ohio, Virginia, Nevada and Pennsylvania. In public polls right now, Romney is losing in each of those states. But, Morris said, that's because the data are all wrong.

"People need to understand that the polling this year is the worst it has ever been," said the onetime Clinton svengali turned conservative pundit.

(See also: POLITICO's poll tracker)

Conservative talk radio king Rush Limbaugh senses something more nefarious in the media and university polls.

"They are designed to do exactly what I have warned you to be vigilant about and that is to depress you and suppress your vote," Limbaugh told his listeners last week, after bringing up swing-state surveys from The Washington Post and CBS News and Quinnipiac University. "These two polls today are designed to convince everybody this election is over."

There has always been a divide between the Republican consultant class and conservative media figures. Operatives must dwell in the real world because their jobs depend on winning and losing. The likes of Morris and Limbaugh have different incentives. They want to build their email lists and listening audiences and there's no faster way to conservative hearts than to kick the dreaded mainstream media. And when it's well after Labor Day of a presidential year and the Republican nominee isn't faring well, reassuring the home team that there's just a scoreboard malfunction offers a seeming dose of logic to the situation.

But the Internet has let the alternate campaign reality flower this fall in a way that's both striking and depressing to political professionals and pollsters. One website, unskewedpolls.com, even readjusts the public polling to include more Republicans in samples. The results: Romney leads in nearly every national poll released in September. Of course arbitrarily reweighting polls is wildly unscientific, but that hasn't stopped Republicans like Limbaugh and even Texas Gov. Rick Perry from mentioning the site.

"Always nice to get unfiltered, or in this case 'unskewed,' information," Perry tweeted last week with a link to the page.

Andrew Kohut, President of the Pew Research Center and before that the head of Gallup, said it's a matter of simple denial.

"We've moved from a place where it looked like the race was close, and there's now an unwillingness to accept reality," said Kohut. "So if you don't like the message you shoot the messenger."

The attempt to debunk polls is in many ways the logical, if absurd, outgrowth of a choose-your-own-adventure political news environment where partisans have outlets that will echo their views.

"It's 'you-are-right' news," sighed longtime Republican strategist Mike Murphy, adding that if it was Democrats losing liberals would be baying about "Rupert [Murdoch] doing mind control on the numbers."

But it's one thing to advocate for a preferred party and portray the other in the worst possible light. Questioning the integrity of professional pollsters is something different: It's preposterous.

Consider: For the vast polling conspiracy of 2012 to be legitimate would be to presume that longtime GOP pollster Bill McInturff is on the deal. McInturff co-runs the respected Wall Street Journal/NBC News poll with veteran Democratic pollster Peter Hart.

McInturff is also business partners with Neil Newhouse, Romney's own pollster. So, by this standard, Romney's own campaign could also be part of the conspiracy to … hurt the Romney campaign.

Of course, there's no such motivation on the part of any pollster hired by mainstream news organizations to gauge public opinion.

"The Prime Directive of pollster survival is to make sure you 'get it right' — whether that be good or bad — for your party," McInturff explained in an e-mail.

There are legitimate discussions in the political community about the partisan breakdown of the recent public polls and whether pollsters should begin to weight data with party preference.

But the recent results that have sparked this wave of conspiracy-mongering are due not to intentional rigging but to the different ways surveys are now conducted.

"Polling has become very complex, with most variation between different pollsters a function of the sample — what percentage of respondents are cell phone only, what percent are 18 to 29 years old or the total percent under or over 45 years old, and the percentage of white versus non-white respondents," noted McInturff.

For example, the automated Rasmussen polls, which have tended to favor Romney, don't contact voters on cell phones, which makes it harder to reach younger voters.

The Romney campaign has repeatedly pointed to the Rasmussen and the Gallup tracking polls in recent weeks to rebut the wave of state-level surveys that have shown Obama opening up a lead.

At a Sept. 18 fundraiser in Utah, Romney finance director Spencer Zwick touted Rasmussen's numbers in the face of mounting data showing a grimmer picture. "We're seeing the polls tighten in key states," Zwick said. The day before, senior Romney strategist Stuart Stevens played down reports of campaign infighting by telling POLITICO that Rasmussen and Gallup told a reassuring story: "If [Obama's] losing 4 points [in Gallup and Rasmussen] and that's a good week, I'd hate to see a bad week."

Obama has since ticked up in the Gallup tracking poll, to a lead in the mid-single digits, while Rasmussen continues to show a nearly deadlocked race.

Most pollsters don't sample according to any set distribution of Republicans, Democrats and independents because partisan identification is not a set attribute like age, race and sex but rather something survey takers are trying to measure to gauge the public's political sentiments at the moment.

"If you look at our party ID trend in 2006, we had Democrats outweighing Republicans 49-41," recalled Kohut. "In 2010, it was only a 47 to 43 margin. If we had adjusted that, we would've missed the boat on how Republican's won."

For this to be questioned has exasperated some longtime pollsters.

"Party identification is basically an attitudinal variable, not a stable population parameter. It is designed to vary," wrote Gallup chief Frank Newport last week in a piece titled "The Recurring — and Misleading — Focus on Party Identification."

Yet some Republicans, seeing some polls with a partisan identification breakdown unlikely to resemble the make-up of a state, believe it's time to start weighting by party.

Veteran GOP pollster Whit Ayres said his preference would be for survey takers to weight their data according to an average of the party ID results they get over the course of an election cycle, in order to minimize variation.

"The poor journalists who get this data are in the position of having to explain why Obama went up by 7 in this month and down by 7 in that month, even though nothing happened and they know in their gut that nothing happened," said Ayres, who added, "I don't buy this sinister motivation that some people attribution to polls. I just think it's the luck of the draw when you're drawing a sample."

Ayres argued there's no reason to avoid reality-based polling, which he said still shows a winnable race: "Are things not going as well as they should for the Romney campaign? Of course, that's just obvious. But we're at the start of the fourth quarter and the guy is a field goal behind."

GOP pollster David Winston, whose firm produced a poll recently showing Obama with a 2-point lead nationally, said pollsters should strive to explain their results to the public in the setting of the partisan sample their surveys turn up: "When you come back with results, you should make sure that you are effectively explaining that ballot test in the context of the demographics contained within the survey."

Asked if he could recall another cycle when one side had so consistently refused to believe unwelcome polling results, Winston recalled the 1998 election, when congressional Republicans were stunned on Election Day to lose seats in the House of Representatives.

"You go back to '98, where Republicans were just convinced that likely voters were showing where the electorate was at, and so you had some saying — and I was not one of them — that this reinforces that Republicans are going to pick up 20 to 30 seats," Winston said. "It turned out not to be the universe that actually showed up."

That was, however, before the Internet became the force it is now. Today, people can seek out polls that favor their side or even find a tonic in the arbitrary rejiggering of professional polls.

"This is the world now," said Murphy. "Just talk to doctors. They all spend half their time explaining themselves to patients who looked stuff up on the Internet and think their doctor is wrong."

© 2012 POLITICO LLC

More information & opt-out options »

What is interest based advertising »

Learn more about quadrantONE »

Privacy Controls by Evidon

[ X ]

This ad has been matched to your interests. It was selected for you based on your browsing activity.

DoubleClick helped quadrantONE determine that you might be interested in an ad like this.

Toyin Falola
Department of History
The University of Texas at Austin
104 Inner Campus Drive
Austin, TX 78712-0220
USA
512 475 7224
512 475 7222 (fax)
http://www.toyinfalola.com
www.utexas.edu/conferences/africa
http://groups.google.com/group/yorubaaffairs
http://groups.google.com/group/USAAfricaDialogue

--
--
You received this message because you are subscribed to the "USA-Africa Dialogue Series" moderated by Toyin Falola, University of Texas at Austin.
For current archives, visit http://groups.google.com/group/USAAfricaDialogue
For previous archives, visit http://www.utexas.edu/conferences/africa/ads/index.html
To post to this group, send an email to USAAfricaDialogue@googlegroups.com
To unsubscribe from this group, send an email to USAAfricaDialogue-
unsubscribe@googlegroups.com

--
There is enough in the world for everyone's need but not for everyone's greed.

---Mohandas Gandhi

--
--
You received this message because you are subscribed to the "USA-Africa Dialogue Series" moderated by Toyin Falola, University of Texas at Austin.
For current archives, visit http://groups.google.com/group/USAAfricaDialogue
For previous archives, visit http://www.utexas.edu/conferences/africa/ads/index.html
To post to this group, send an email to USAAfricaDialogue@googlegroups.com
To unsubscribe from this group, send an email to USAAfricaDialogue-
unsubscribe@googlegroups.com

Dialogues

Monday, October 1, 2012

Re: USA Africa Dialogue Series - Can Bangura be right?

No comments:

Post a Comment

Followers

Blog Archive

About Me


The parallel universe where Mitt Romney leads all polls By: Jonathan Martin and Alexander Burns October 1, 2012 04:27 AM EDT
To talk with any working Republican political operative these days is to hear the same tale of woe: a grim accounting of the past few weeks, a dash of gallows humor and a measure of hope that President Obama is still beatable. Never in question is that Mitt Romney is trailing — the private surveys these strategists see for their down-ballot clients make that clear. The only question is by how much. But hanging up the phone or clicking out of e-mail is to find a parallel universe on the right. On TV, talk radio and especially the Internet is a place where the swing-state polls that show Romney losing are not just inaccurate but part of an intentional plot by the heretofore unknown media-pollster axis to depress Republican voters. In this other world, Romney not only isn't losing — he's on the verge of a convincing victory. "I believe if the election were held today, Romney would win by 4 or 5 points," trumpeted Dick Morris on Fox News last week, predicting a win for the GOP ticket in Florida, Ohio, Virginia, Nevada and Pennsylvania. In public polls right now, Romney is losing in each of those states. But, Morris said, that's because the data are all wrong. "People need to understand that the polling this year is the worst it has ever been," said the onetime Clinton svengali turned conservative pundit. (See also: POLITICO's poll tracker) Conservative talk radio king Rush Limbaugh senses something more nefarious in the media and university polls. "They are designed to do exactly what I have warned you to be vigilant about and that is to depress you and suppress your vote," Limbaugh told his listeners last week, after bringing up swing-state surveys from The Washington Post and CBS News and Quinnipiac University. "These two polls today are designed to convince everybody this election is over." There has always been a divide between the Republican consultant class and conservative media figures. Operatives must dwell in the real world because their jobs depend on winning and losing. The likes of Morris and Limbaugh have different incentives. They want to build their email lists and listening audiences and there's no faster way to conservative hearts than to kick the dreaded mainstream media. And when it's well after Labor Day of a presidential year and the Republican nominee isn't faring well, reassuring the home team that there's just a scoreboard malfunction offers a seeming dose of logic to the situation. But the Internet has let the alternate campaign reality flower this fall in a way that's both striking and depressing to political professionals and pollsters. One website, unskewedpolls.com, even readjusts the public polling to include more Republicans in samples. The results: Romney leads in nearly every national poll released in September. Of course arbitrarily reweighting polls is wildly unscientific, but that hasn't stopped Republicans like Limbaugh and even Texas Gov. Rick Perry from mentioning the site. "Always nice to get unfiltered, or in this case 'unskewed,' information," Perry tweeted last week with a link to the page. Andrew Kohut, President of the Pew Research Center and before that the head of Gallup, said it's a matter of simple denial. "We've moved from a place where it looked like the race was close, and there's now an unwillingness to accept reality," said Kohut. "So if you don't like the message you shoot the messenger." The attempt to debunk polls is in many ways the logical, if absurd, outgrowth of a choose-your-own-adventure political news environment where partisans have outlets that will echo their views. "It's 'you-are-right' news," sighed longtime Republican strategist Mike Murphy, adding that if it was Democrats losing liberals would be baying about "Rupert [Murdoch] doing mind control on the numbers." But it's one thing to advocate for a preferred party and portray the other in the worst possible light. Questioning the integrity of professional pollsters is something different: It's preposterous. Consider: For the vast polling conspiracy of 2012 to be legitimate would be to presume that longtime GOP pollster Bill McInturff is on the deal. McInturff co-runs the respected Wall Street Journal/NBC News poll with veteran Democratic pollster Peter Hart. McInturff is also business partners with Neil Newhouse, Romney's own pollster. So, by this standard, Romney's own campaign could also be part of the conspiracy to … hurt the Romney campaign. Of course, there's no such motivation on the part of any pollster hired by mainstream news organizations to gauge public opinion. "The Prime Directive of pollster survival is to make sure you 'get it right' — whether that be good or bad — for your party," McInturff explained in an e-mail. There are legitimate discussions in the political community about the partisan breakdown of the recent public polls and whether pollsters should begin to weight data with party preference. But the recent results that have sparked this wave of conspiracy-mongering are due not to intentional rigging but to the different ways surveys are now conducted. "Polling has become very complex, with most variation between different pollsters a function of the sample — what percentage of respondents are cell phone only, what percent are 18 to 29 years old or the total percent under or over 45 years old, and the percentage of white versus non-white respondents," noted McInturff. For example, the automated Rasmussen polls, which have tended to favor Romney, don't contact voters on cell phones, which makes it harder to reach younger voters. The Romney campaign has repeatedly pointed to the Rasmussen and the Gallup tracking polls in recent weeks to rebut the wave of state-level surveys that have shown Obama opening up a lead. At a Sept. 18 fundraiser in Utah, Romney finance director Spencer Zwick touted Rasmussen's numbers in the face of mounting data showing a grimmer picture. "We're seeing the polls tighten in key states," Zwick said. The day before, senior Romney strategist Stuart Stevens played down reports of campaign infighting by telling POLITICO that Rasmussen and Gallup told a reassuring story: "If [Obama's] losing 4 points [in Gallup and Rasmussen] and that's a good week, I'd hate to see a bad week." Obama has since ticked up in the Gallup tracking poll, to a lead in the mid-single digits, while Rasmussen continues to show a nearly deadlocked race. Most pollsters don't sample according to any set distribution of Republicans, Democrats and independents because partisan identification is not a set attribute like age, race and sex but rather something survey takers are trying to measure to gauge the public's political sentiments at the moment. "If you look at our party ID trend in 2006, we had Democrats outweighing Republicans 49-41," recalled Kohut. "In 2010, it was only a 47 to 43 margin. If we had adjusted that, we would've missed the boat on how Republican's won." For this to be questioned has exasperated some longtime pollsters. "Party identification is basically an attitudinal variable, not a stable population parameter. It is designed to vary," wrote Gallup chief Frank Newport last week in a piece titled "The Recurring — and Misleading — Focus on Party Identification." Yet some Republicans, seeing some polls with a partisan identification breakdown unlikely to resemble the make-up of a state, believe it's time to start weighting by party. Veteran GOP pollster Whit Ayres said his preference would be for survey takers to weight their data according to an average of the party ID results they get over the course of an election cycle, in order to minimize variation. "The poor journalists who get this data are in the position of having to explain why Obama went up by 7 in this month and down by 7 in that month, even though nothing happened and they know in their gut that nothing happened," said Ayres, who added, "I don't buy this sinister motivation that some people attribution to polls. I just think it's the luck of the draw when you're drawing a sample." Ayres argued there's no reason to avoid reality-based polling, which he said still shows a winnable race: "Are things not going as well as they should for the Romney campaign? Of course, that's just obvious. But we're at the start of the fourth quarter and the guy is a field goal behind." GOP pollster David Winston, whose firm produced a poll recently showing Obama with a 2-point lead nationally, said pollsters should strive to explain their results to the public in the setting of the partisan sample their surveys turn up: "When you come back with results, you should make sure that you are effectively explaining that ballot test in the context of the demographics contained within the survey." Asked if he could recall another cycle when one side had so consistently refused to believe unwelcome polling results, Winston recalled the 1998 election, when congressional Republicans were stunned on Election Day to lose seats in the House of Representatives. "You go back to '98, where Republicans were just convinced that likely voters were showing where the electorate was at, and so you had some saying — and I was not one of them — that this reinforces that Republicans are going to pick up 20 to 30 seats," Winston said. "It turned out not to be the universe that actually showed up." That was, however, before the Internet became the force it is now. Today, people can seek out polls that favor their side or even find a tonic in the arbitrary rejiggering of professional polls. "This is the world now," said Murphy. "Just talk to doctors. They all spend half their time explaining themselves to patients who looked stuff up on the Internet and think their doctor is wrong."
© 2012 POLITICO LLC