Lack of Iowa Poll is disappointing but not a conspiracy

February 3, 2020 6:00 am

The Des Moines Register announced it would not publish its much-anticipated Iowa Poll on Feb. 2, 2020, after a polling error. (Photo by Kathie Obradovich/Iowa Capital Dispatch)

I have a lapel pin that looks like a tiny newspaper front page with the headline:  “Dewey defeats Truman.”

Of course, we know from history that Harry Truman won the presidency in 1948, not Thomas Dewey.  I think about that premature headline on big election days as a simple reminder:  Don’t make stupid predictions.

Going into caucus day without a Des Moines Register Iowa Poll to lean on makes that both easier and harder.

I was as shocked as anyone in the political world when the Register announced Saturday night that it would not release its much-anticipated final poll before the Iowa caucuses.  I also felt a sharp sense of grief, almost like I was learning about a sudden death in the family.

Most of that came from empathy for my former colleagues at the Register and for pollster Ann Selzer, who must have been devastated.  The Register reported that the decision was made to cancel publication of the poll after Pete Buttigieg’s campaign complained that his name had been left off the list of candidates by an interviewer during at least one survey call.

CNN, the Register’s partner for the poll, did an internal investigation that found that an operator had enlarged the font on his or her computer screen. That cut off Buttigieg’s name, which was at the end of the list, the Register reported.  The list of candidates is randomized so the names are in a different order with each interview. The Register reported that the polling partners were unable to determine whether this error happened more than once to Buttigieg or other candidates.

The Register, CNN and Selzer made an excruciatingly hard decision, but it was the right one.  They could not be sure of the integrity of the poll and integrity was the only thing that mattered.

That was a disappointment to campaign operatives who thought their candidate might have been leading the poll.  That’s understandable.  While polls aren’t predictions, there was certain to be a positive wave of media for the candidate in the lead just as many caucusgoers were making up their minds.

There were also a few uninformed and frankly irresponsible conspiracy theories aired on social media, suggesting the media partners would suppress poll results showing a certain candidate was winning. Never in the 16 years I worked at the Register, including working directly with this poll, was there ever a moment’s consideration given to any concern but whether the poll was conducted and presented accurately and fairly.

There’s simply no question that the cost — in money, time and reputation — for withholding this poll was far greater than any imaginary political benefit.  The decision, as disappointing as it was for everyone, should actually enhance the Iowa Poll’s legendary reputation, not tarnish it.

I admit that part of my emotional reaction to not having a poll also meant that I will not be as prepared to answer questions about what people can expect on caucus night. It’s kind of a security blanket for media to have a trustworthy poll to not only suggest a potential outcome but also to give us some ready answers for why people are making their choices.

Having said all that, however, the lack of a Register poll adds to the perception that just about anything could happen on caucus night.  And that’s not such a bad thing. It should be voters, not the media, who pick the winners and losers. There has been too much emphasis over the years on whether a candidate beat media-set expectations instead of whether he or she beat the other contenders.

The fact that this or that candidate may be polling better than another in the days ahead of the caucuses shouldn’t matter to Iowa voters. Not a single vote can be tallied until they show up.

I’m looking forward to waiting to see what voters decide.







Our stories may be republished online or in print under Creative Commons license CC BY-NC-ND 4.0. We ask that you edit only for style or to shorten, provide proper attribution and link to our web site. Please see our republishing guidelines for use of photos and graphics.