Two polls paint different picture of presidential race in Iowa

Even with perfect methodology–random sampling, neutral question wording, and an accurate likely voter screens–approximately one out of every 20 political polls will be wrong outside the margin of error, just by chance. Two polling firms released snapshots of the Iowa electorate yesterday. They can’t both be right.

The NBC News/Wall Street Journal/Marist snapshot of the Iowa electorate will suit Democrats. The ticket of President Barack Obama and Vice President Joe Biden leads Republican Mitt Romney and Representative Paul Ryan by 50 percent to 42 percent among likely voters and by 51 percent to 40 percent among registered voters. Mark Murray summarizes more of the toplines here. Obama’s approval rating is 49 percent in Iowa, and his favorability ratings are positive (53/42), while Romney’s are negative (42/50). By a 47 percent to 43 percent margin, respondents think Obama would do a better job handling the economy.

Full results from the NBC/Wall Street Journal/Marist poll are here (pdf). The document includes results from the same firm’s last Iowa poll in May, which showed Obama and Romney tied at 44 percent. The likely voter sample for the new poll consisted of 898 Iowans surveyed from September 16 to 18, producing a margin of error of plus or minus 3.3 percentage points. The margin of error for the registered voter sample of 1,253 Iowans was plus or minus 2.8 percent.

Among both likely and registered voters, NBC/WSJ/Marist found that 43 percent of respondents think the country is headed in the right direction, while 49 percent think it’s off on the wrong track. That’s an improvement from the May survey, in which 39 percent said “right direction” and 54 percent said “wrong track.” Still, one would not expect a sitting president to be leading his challenger outside the margin of error when 49 percent of the sample believes the country’s on the wrong track.

Rasmussen Reports’ latest Iowa paints a better picture for Republicans:

The latest Rasmussen Reports telephone survey of Likely Iowa Voters shows Romney with 47% support to President Obama’s 44%. Three percent (3%) prefer some other candidate in the race, and five percent (5%) are still undecided.

The margin of error for that survey of 500 Iowans on September 19 is plus or minus 4.5 percent. I can’t post more toplines or other details, because I am not a Rasmussen subscriber. This poll suggests that the president did not receive any “bounce” from the Democratic National Convention and other news of the past two weeks. In August, Rasmussen Reports found Romney ahead in Iowa by 46 percent to 44 percent.

In past election years, the Des Moines Register has had Selzer & Co. conduct a statewide poll in mid-September, and another poll the week before the November election. I don’t know the Register’s timetable for polling this fall.

Any comments about the presidential election are welcome in this thread. New Jersey Governor Chris Christie gave a pep talk to an audience of Republican donors in Altoona last night, a few hours after raising money for Representative Steve King in Sioux City.

Ann Romney headlined an event in Clive yesterday to showcase her husband’s “good side,” which “very few people know about.” Radio Iowa’s O.Kay Henderson posted the audio of her speech and an exclusive interview. This part will go over well, I’m sure.

Mrs. Romney directly addressed her fellow Republicans who’ve criticized her husband.

“Stop it. This is hard. You want to try it? Get in the ring,” she said. “This is hard and, you know, it’s an important thing that we’re doing right now and it’s an important election and it is time for all Americans to realize how significant this election is and how lucky we are to have someone with Mitt’s qualifications and experience and know-how to be able to have the opportunity to run this country.”

The criticism and very public hand wringing from Republicans comes as national polls show the race between Romney and President Obama is essentially tied.

“It’s nonsense and the chattering class…you hear it and then you just let it go right by,” she told Radio Iowa. “…Honestly, at this point, I’m not surprised by anything.”

About the Author(s)

desmoinesdem

  • Rasmussen

    His polls are out of whack with other pollsters in a lot of states. Doesn’t mean he’s wrong, just that he’s probably wrong.

  • the key difference

    is found here:

    NBC:

    Likely voters (Sept 2012):

    D 36

    R 31

    NP 33

    Actual:

    2008

    D 37

    R 32

    NP 31

    2004

    D 33

    R 35

    NP 32

    Essentially, they are saying that turnout looks more like a 2008 model with even poorer performance for Republicans.  

    • This is why the NBC poll is not as accurate

      With Republican registration up in Iowa, Obama’s favorability down, a repeat of 2008 numbers would be nearly impossible.

      Iowa is probably pretty close to a coin flip right now.

      • I give the edge to Obama

        partly because he seems to have gained a few points almost everywhere the past few weeks, partly because Democrats are likely to go into election day with an early voting advantage. But I doubt he will win by an 8-9 point margin.  

        • I would concur

          Pretty close to a coin flip, but not quite there.  Obama does have the edge, but I’d be very surprised if it’s outside the 1 to 3 point range.

Comments