Subsequently, NewsBusters' Noel Sheppard picked up on how “Tom Bevan of Real Clear Politics posted an analysis of this poll’s methodology at his blog last evening. What his figures show is that CBS polled 46% more Democrats in its weighted sample than Republicans.” In short, CBS polled 259 Republicans, but weighted the sample to count for only 223, or 24 percent of the total; they surveyed 326 Democrats and held that number so they represented 35 percent of those polled; and independents moved from 351 respondents to a weighted 388 for 41 percent of the sample. Sheppard explained: “To put these numbers in proper perspective, according to the November 2004 exit polls, the nation’s current party affiliation is 37 percent Democrats, 37 percent Republicans, and 26 percent independents. As such, the polling agency involved in this result fell 36 percent short in sampling Republicans while over-sampling independents by 59 percent.” (More on Brit Hume's "Grapevine" item on the poll, weighting and the Borger story in full, follows.)
Vaughn Ververs, Editor of CBS News' “Public Eye” blog, in an item which cited the Wednesday NewsBusters/MRC CyberAlert take on Roberts/Nixon, took up the weighting revelation by Real Clear Politics. First, in his November 3 item titled, “Bad News Isn’t Fun But It Doesn't Last Forever,” he cited a general explanation of weighting which is posted on the CBSNews.com site:
“At the end of our surveys, we find sometimes that we have questioned too many people from one group or another. Older people, for example, tend to be at home to answer the phone more than younger people, so there is often a greater percentage of older people in our surveys than exists in the American public.”
He then asked:
“So how does a poll end up being comprised of 35% Democrats, 41% Independents and 24% Republicans? [Director of Surveys for CBS News Kathy] Frankovic explains that there is no adjusting for party identification in the CBS News polls and that the weighting described above accounts for the changes in the party ID. Frankovic notes: 'The people who tend to be under-represented in a sample...tend to be younger and tend to be more minorities. So that by assuring that the sample looks like the country, you’re probably going to almost always increase the number of Democrats.' In short, when the sample is adjusted to match the Census Bureau data, the party ID percentages change.”
That still doesn't explain for me the much higher percent of Democrats and independents counted than Republicans which does not match their percentages in the population, a skew that I think is evidenced in how CBS's 35 percent approval for Bush is four or more points lower than other recent polls.
For the finding Borger cited, check here for page 2 of the CBSNews.com posting of the poll released Wednesday. Scroll down to: “COMPARED TO MOST PEOPLE IN PUBLIC LIFE, BUSH'S HONESTY & INTEGRITY IS...”
Go here for the PDF from CBS News of all the poll numbers, with the “unweighted” versus “weighted” numbers by party affiliation at the very end.
Now, Borger's November 3 CBS Evening News story in full, which was followed by anchor Bob Schieffer wanting to know if Karl Rove will “have to leave?”
Gloria Borger: “By all appearances, it's business as usual for Rove, but Republican sources say his problems are a distraction even for this disciplined administration. Ron Klain was Vice President Al Gore's chief of staff throughout much of the turmoil in the Clinton years and says disruption is inevitable.”
Ron Klain: “It would be inhuman not to be distracted or diverted by reading in the newspaper every day that your colleagues are speculating about your departure.”
Borger: “And tension creeps into the West Wing when staffers have to testify about each other under oath.”
Klain: “It complicates White House communications at the very least.”
Borger: “Just ask Scott McClellan, the President's Press Secretary, who said this about 'Scooter' Libby and Karl Rove two years ago.”
McClellan, October 10, 2003: “I spoke with those individuals, as I've pointed out, and those individuals assured me they were not involved in this.”
Borger: “That inaccuracy haunts McClellan-”
Voice of NBC's David Gregory to McClellan, unknown date: “You were wrong then, weren't you?”
Borger: “-and affects the President.”
Borger: “A new CBS News poll shows that only 39 percent of Americans say that President Bush has more honesty and integrity than most people in public life, down eleven points since early last year [on screen: “about the same” at 27 percent, “less” at 29 percent”]. White House advisers are supposed to provide reality checks for Presidents isolated in the Oval Office.”
Klain: “Then he becomes completely dependent on what filters to him. I think White House aides are no different than any employee in any office in any place in America. Which is, you want to tell the boss good news and you want to be on the boss' good side, you hate to tell him he is messing up.”
Borger: “But someone has to do it and that's why some inside the White House now are quietly lobbying for change, Bob.”
Schieffer: “Well, what do you hear? This is a White House, Gloria, that as we all know that doesn't say much. But there's a lot of talk about this. Where do you think they're coming down on this. Is he going to have to leave?”
Borger: “Well, what's fascinating about this, Bob, as you point out, this is a White House that doesn't really leak. And now we see a White House that's very publicly divided. There's the pro-Karl Rove camp, the people that think the White House cannot function without him. There's the anti-Karl Rove camp, the people who believe that he can no longer a be effective and needs to leave if the President is going to get beyond this CIA investigation.”
