Iowa. (Chip Somodevilla/Getty Images)

J. Ann Selzer may be the single most powerful pollster in America. She runs the Des Moines Register's Iowa poll, the gold standard of survey research in what is the single most-watched state, politically speaking, in the country. So when Selzer's final numbers in the heated Senate race between state Sen. Joni Ernst (R) and Rep. Bruce Braley (D) were released Saturday night, they caused quite the shock wave on the political scene. The poll showed Ernst up seven points — 51 percent to 44 percent — and immediately sent the Braley team into a funk.  After the initial shock wore off, they began to attack Selzer's numbers. I reached out to her to talk about the poll, the reaction to the poll and what it's like to poll in Iowa. Our conversation, edited only for grammar, is below.

FIX: Let's start with the poll.  What — inside the numbers — produced a larger lead for Ernst than some other polls have shown?

Selzer: I have not seen others' poll in detail, but we had Ernst lead[ing] in Braley's home district and we had her winning independents.  Not sure other polls showed that.  We looked specifically at the claim that Braley's organization was identifying erratic mid-term voters.  Ernst won with respondents who say they voted in 2010 and those who did not.  I have no idea how they are modeling their data.

FIX: Were you surprised at the blowback to the poll release? One aide to Braley said, "I think this could be a real big credibility issue for the Register and Ann Selzer."

Selzer: I've heard it before. What else could they say?  All I can do as a pollster at this point is wait and worry.  We'll know soon enough. [Editor's note: In 2008, Selzer took massive amounts of criticism for her final Register poll that showed Barack Obama running away with the Iowa caucuses. And we know how that turned out.]

FIX: One thing I was surprised by in the poll was how much Braley's pejorative comments about farmers had hurt him.  Can you explain, from the numbers, why that is — and with what groups it hurt the most?

Selzer: Take a look at the exit polls for both [Sen. Tom] Harkin in 2008 and [Sen. Chuck] Grassley in 2010.  They both run the demographic tables.  They each won by about the same large margin.  Grassley got 23 percent of the Democratic vote.  This means there are a lot of crossover voters who voted for Harkin and for Grassley.  It's hard to say Grassley isn't qualified to chair the judiciary committee, given he has served on it for a long time.  This is a case, in my opinion, that the comment was bad, but not fixing it — thinking this would blow over — was the bigger mistake.

FIX: How has polling Iowa changed in the time you have been doing it? In the same ways polling nationally has changed — more cells, lower response rates, etc. — or different ways?

Selzer: There is so much more to worry about.  We've addressed the cellphone issue and we no longer consider it to be the problem it once was.  The low response rates worry me, though Pew did a wonderful study showing that they are not as big a problem as we might think.  This is a problem, however, that will grow.  As money pours into a small state like Iowa, voters' phones are ringing off the hook.  I worry a lot about a time when people who respond to polls now [who] will get fed up.

FIX:  Multiple choice. Iowa a) leans slightly Democratic in most elections b) leans slightly Republican in most elections or c) is a true tossup. And, why?

Selzer: Iowa takes the measure of the candidate.  Sometimes a Democrat (like Harkin, former governor Tom Vilsack, etc.) feels best.  Sometimes a Republican (Grassley, Gov. Terry Branstad) feels best.  Those specific races (with the exception of Vilsack's first win) were not tossups.  So I'm not sure it tilts slightly in these races.  It just seems Iowans are willing to put party aside and vote for the candidate the voter likes best across a number of dimensions. Honestly, that's one of the things that makes Iowa a fun state to poll.