May 13, 2013 09:05 AM UTC

Gallup Explains What Went Wrong, Except The Obvious

•  
• by: Colorado Pols

Politico:

After misreading the 2012 presidential election and facing criticism in the aftermath, Gallup polling has undertaken an internal review and will announce the findings next month.

“We are in the process of finishing a full review of all methodological issues relating to our 2012 election polling. The process is being led by a blue-ribbon group of outside experts. We will be reporting our findings at an event on June 4 at our offices in Washington,” Gallup Editor-in-Chief Frank Newport told POLITICO’s Mike Allen in Playbook on Monday.

Once considered a gold standard in public polling, Gallup's failure to correctly forecast the winner of the 2012 presidential election, and consistent track to the right of what turned out to be voter sentiment last year, have hurt their credibility in much the same way as CNN's recent string of "breaking" misreports on the Boston bombing have tarnished what was once an inviolate reputation.

There's a larger context to Gallup's mistaken assumptions about "likely voters" in the last election, and Gallup was far from the only "credible" news source whose presumptions about the 2012 electorate were proven wrong. Based on assumptions that serviced partisan desires, datasets skewed by those assumptions, and a massive nine-figure marketing campaign, a very large number of Americans bought into an artificial and ultimately false sense of certain GOP victory. From talk-radio listener to pollster, assumptions were made first, then self-serving datasets were used to reinforce them. We called all of this as it happened, and were ourselves accused of partisanship.

Fixing the mechanics of Gallup's polling will be the easy part.