Topic: Media Research Center
Sometimes the Media Research Center's right hand doesn't know what its ... er, other right hand is doing.
A Sept. 10 NewsBusters post by Matthew Balan complained that a CNN "poll on President Obama’s health care speech to Congress on Wednesday significantly oversampled Democrats." Balan later appended a statement from MRC chief Brent Bozell:
CNN has no business airing a ‘poll’ that is straight partisan reaction from those already in the can for ObamaCare. Caveat or no caveat, CNN was dead wrong to run this and on national television proclaim, ‘We have important numbers here.’ It is an embarrassment that their news directors deemed it newsworthy to begin with. This network simply cannot be trusted to cover this hugely important story fairly.
But Balan and Bozell overlooked on crucial aspect of the poll -- it sampled only people who viewed Obama's speech. Balan did note that a CNN article "didn’t mention the over-representation of Democrats until the third paragraph," which stated that the poll "surveyed the opinions of people who watched Wednesday night's speech, and does not reflect the views of all Americans."But Balan didn't note this until the 10th and final paragraph of his original post.
It fell on another MRC division to prove Balan and Bozell wrong. A Sept. 11 CNSNews.com article by Pete Winn not only accurately describes the poll's methodology but does so by quoting a conservative pollster, Kellyanne Conway:
But Conway, who has questioned the validity of numerous polls in the past for being slanted against conservatives, said it would be wrong to question the validity of this CNN poll.
“This is a poll of debate-watchers – not the general population,” the pollster said.
In fact, polls like this one are not designed to sample what the American public at large thinks – they are “a random-call measure of who the audience really was – and what those specific people thought,” she told CNSNews.com.
“To their (CNN’s) enormous credit, they reported it that way,” Conway told CNSNews.com.
“They did not say ‘2 out of 3 Americans now support the president on health-care reform,” Conway said. “What (CNN) said, from the beginning, was ‘2 out of 3 Americans who watched the speech said . . .’”
In this case, she said, Wednesday night’s poll talked to more Democrats simply because Obama’s audience itself leaned overwhelmingly Democratic.[...]
In fact, CNN “went the extra mile” to make it clear that the number of those who identified themselves as Democrats was “8 to 10 percentage points higher than the national Democratic sample would be if this were a survey of voters,” Conway said.
[...]
“I can bet the farm, there will be pundits and programs and hosts, who will say, ‘According to the CNN poll, the president has really gotten back on track. Two-thirds of Americans now support his health-care. . .’ No-no. They would be misreading the poll and using the poll for their own purposes.
“But in fairness, CNN did not report it that way,” she added.
Winn even quoted a CNN spokesperson to back it up:
“We can’t control who watched the speech, but we found those people using scientific methods,” a CNN executive, who asked not to be identified, said.
The number of Republicans versus Democrats in the sample will vary depending on which party is in office, the spokesman said.
“It goes back and forth, and back and forth depending on which party controls the White House, but we’ve seen this for a really long time,” the spoksman added.
“In 2005, just after the 2005 State of the Union speech, for example, 52 percent of our sample were Republicans – that’s about 16 points higher than the whole public at that time -- and only 25 percent of the speech-watchers were Democrats, and that’s about 7 points under what you might have expected at that time if you had done a sample of the whole population.”
The analogy is to baseball, according to CNN.
“The audience (for a presidential speech) tends to self-select, like a Yankee audience and a Red Sox audience,” the spokesman said. “Red Sox fans don’t watch Yankee games and Yankee fans don’t watch Red Sox games.”
Winn gets extra points for accurately identifying Conway as a conservative pollster, something CNS has failed to do in the past. But he loses points for writing that "some conservatives in Washington were buzzing about the poll’s sample – and questioning whether it might have been partisan or slanted" and not noting that one of those conservatives was his boss.
So, will Winn's story spread across the rest of the MRC empire, or do Balan and Bozell care more about maintaining their fictional liberal-media-bias tunnel vision than they do about the truth?
