Perhaps you stumbled across news yesterday that Gov. Pawlenty scored a 56 percent “positive job rating,” as the Strib put it, compared to 41 percent for Sen. Al Franken.
Rasmussen Reports did the survey, which got play in the Strib’s paper version and in the PiPress’ Political Animal blog, among others. PiPresser Bill Salisbury noted that Pawlenty’s 56 percent “approval” matched U.S. Sen. Amy Klobuchar’s.
But here’s the thing about comparing the numbers: Rasmussen didn’t ask the same question about the Republican governor as it did about the Democratic senators. And that likely makes Franken and Klobuchar look less popular, relatively.
Here’s the poll’s question about TPaw:
“How would you rate the job Tim Pawlenty has been doing as Governor… do you strongly approve, somewhat approve, somewhat disapprove, or strongly disapprove of the job she’s been doing?”
Gender-bender aside — we can only hope “she” was a web typo — things breaks cleanly: the first two categories (26 and 30 percent, respectively) indicate approval, and the other two (13 and 30 percent), disapproval.
So it’s accurate to say Pawlenty has a 56 percent approval rating, within the 500-person survey’s plus/minus 4.5 percent margin of sampling error.
For some reason, though, Ras got murkier with the senators. For example: “How would you rate the way that Al Franken is performing his role as Senator….excellent, good, fair or poor?” (The Klobuchar question was identical.)
Surely, those ranking Al or Amy as “excellent” or “good” approve. “Fair”? Not quite so clear.
It’s possible, even likely, that some folks rating the senators “fair” would also say they “somewhat approve” of their performance. Denying Franken and Klobuchar that clarity makes them look worse relative to Pawlenty.
Polling boss Scott Rasmussen has conducted polls for the Republican National Committee and George W. Bush, though lefty pollsters like FiveThirtyEight.com’s Nate Silver have praised his objectivity. But in this case, Rasmussen’s survey design is lacking.