Unfortunately, it looks like a couple of media outlets are painting slightly curious assessments of their own polling. And the common thread, as it happens, is to draw the common conclusion that the Obama bounce wasn't all that bouncy. In one case, it was a bit of a strange sense of timing. In the other, however, it was a criminal misuse of their own statistics.
On to the numbers:
PRESIDENTIAL GENERAL ELECTION TRIAL HEATS:
NATIONAL (ABC/Washington Post): Obama d. Romney (49-48 LV; 50-44 RV)DOWNBALLOT POLLING:
NATIONAL (Gallup Tracking): Obama d. Romney (50-44)
NATIONAL (Ipsos-Reuters Tracking): Obama d. Romney (46-43 LV; 45-39 RV)
NATIONAL (PPP for Daily Kos/SEIU): Obama d. Romney (50-44)
NATIONAL (Rasmussen Reports): Obama d. Romney (48-45)
ARIZONA (PPP for the League of Conservation Voters): Romney d. Obama (53-44)
FLORIDA (SurveyUSA): Obama d. Romney (48-44)
MINNESOTA (SurveyUSA): Obama d. Romney (50-40)
VIRGINIA (Gravis Marketing--R): Romney d. Obama (49-44)
AZ-SEN (PPP for the League of Conservation Voters): Jeff Flake (R) 44, Richard Carmona (D) 43A few thoughts, as always, await you just past the jump...
FL-SEN (SurveyUSA): Sen. Bill Nelson (D) 47, Connie Mack IV (R) 36
OH-SEN (PPP): Sen. Sherrod Brown (D) 48, Josh Mandel (R) 40
VA-SEN (Gravis Marketing--R): George Allen (R) 48, Tim Kaine (D) 43
On today, the Tuesday after the DNC, Gallup issued their "official" statement about their convention bounce. The banner headline atop their website read, curiously, as follows: "Obama Gets Three-Point Convention Bounce."
Regular readers of the Wrap would find that to be strange. After all, at the dawn of the gathering in Charlotte last week, the president led by a margin of a single point (47-46). Today, that margin sits at six points (50-44). That would seem like a five-point bounce. But what Gallup did, to reach their calculation, was to simply measure the change in Obama's support (from 47 percent to 50 percent), rather than the change in margin between Obama and Romney.
The problem there is twofold. For one thing, on a basic level, is the fact that given the outsized polling window utilized by Gallup, the Obama bounce may not necessarily be done. Today's sample was based on polls from Tuesday through Monday. Ergo, over 40% of the respondents came before Obama's convention speech, and over a quarter had not seen Bill Clinton's speech. The second problem is that the "three-point bounce" has been repeated, including in media outlets that have calculated the bounce previously by the difference in the margin. So, the net effect is to make the Gallup bounce seem smaller than it really is.
But that pales in comparison to what ABC/Washington Post did in the reporting of their poll. Check this out:
HEADLINE: "Among likely voters, Obama-Romney close"The grievous act of malpractice there? The poll taken "just before the conventions" was based on a sample of registered voters, not likely voters. Right in the lede, WaPo makes an apples-to-oranges comparison to create a very false equivalence between the two polls.
Last week’s Democratic National Convention helped President Obama improve his standing against Republican Mitt Romney, according to a new Washington Post-ABC News poll, but did little to reduce voter concern about his handling of the economy.
The survey shows that the race remains close among likely voters, with Obama at 49 percent and Romney at 48 percent, virtually unchanged from a poll taken just before the conventions.
But among a wider sample of all registered voters, Obama holds an apparent edge, topping Romney at 50 percent to 44 percent, and has clear advantages on important issues in the campaign when compared with his rival.
The reality of the situation, of course, is dramatically different. President Obama, on the margin, is polling seven points stronger now against Mitt Romney than he did in the pre-convention poll among registered voters. And that is quite the statement: it is not saying that Barack Obama got a lofty seven-point bounce. It is bigger than that: it is saying that the Obama DNC bounce was seven points larger than Romney's RNC bounce, because it clearly more than offset any bump Romney received during his confab in Tampa.
But, if you read the WaPo writeup, it seems like the conventions barely moved the needle. That is a blatant misrepresentation of the actual data, and it would be hard to deny liberal conspiracy theories that a major media outlet is actively trying to distort their own data to create the impression of a stubborn tossup unchanged by the convention season. Shame on them for that.
In other polling news...
- SurveyUSA wasn't kidding when they tweeted earlier this week that they were getting back into the game with a vengeance. In addition to yesterday's (mighty odd) North Carolina poll, they have released data in the last 36 hours in Washington, Minnesota, and Florida. Unlike the bizarre African-American sub-data in the Tar Heel State, the other three states came closer to hitting the fairway in terms how the numbers "look." There were some minor quibbles, but nothing as dramatic as that 30 percent Romney "support" among black North Carolinians.
- PPP hints at another Senate tossup with their latest numbers in Arizona, even as Mitt Romney has a modest-yet-clear lead in the state. Jeff Flake's lead is a single point. Digging into the details, right-wingers will point out that the sample is a little sympathetic to Democrats (the sample went 50-44 for McCain in 2008, when he really won 54-45). Liberals, however, will point out that the sample is pretty darned pale (83 percent white--the 2008 exit poll was only 75 percent white). On balance, it seems like a pretty fair assessment of the state of play, especially when you consider that this is the second poll out of Arizona showing a tossup, with no contrary claims of recent vintage from Jeff Flake or his GOP allies.
- As for that poll out of Virginia, all I will say is this: I am waiting for legitimate polling out of the state before I pass judgment on the state of play in the state. And...as Forrest Gump was wont to say..."That's all I've got to say about that."