In 2019, I wrote a post, “Remember that paper we wrote, The mythical swing voter? About shifts in the polls being explainable by differential nonresponse? Mark Palko beat us to this idea, by 4 years,” where I quoted Palko as writing in 2012:

So response bias was amplified by these factors:

1. the effect was positively correlated with the intensity of support

2. it was accompanied by matching but opposite effects on the other side

3. there were feedback loops — supporters of candidates moving up in the polls were happier and more likely to respond while supporters of candidates moving down had the opposite reaction.

The above completely anticipates the main result of our Mythical Swing Voter paper, which is based on the Xbox polling data we collected in 2012, analyzed in 2013, wrote up in 2014, and published in 2016, and which was picked up in the news media in time for the 2016 campaign.

My 2019 post concludes: “Doug Rivers (one of my coauthors on the Mythical Swing Voter paper) was also talking in 2012 about differential nonresponse; see the last three paragraphs here.”

That’s a link to a news article by Mark Blumenthal. I happened to reread it recently (in preparation for a talk I’m giving called “Wrong Again! 30+ Years of Statistical Mistakes,” about all sorts of errors I’ve made) and I came across this passage from the middle of the article that I hadn’t noticed before:

A second theory, outlined on Wednesday by The New York Times’ Nate Silver, is that polls exaggerate the “bounces” from events like debates that could temporarily “affect voter enthusiasm and their willingness to respond to surveys.”

By this theory, the Democratic convention and the release of the “47 percent” tape boosted Democratic enthusiasm and the first debate boosted Republican enthusiasm, and these changes affected the relative willingness of partisans on either side to respond to telephone surveys. “Perhaps,” Silver speculates, “Mr. Obama’s numbers went from being artificially inflated to artificially deflated, exaggerating the degree of change in the race on both ends.”

So we were scooped by Mark Palko, Doug Rivers, and Nate Silver.

I’m not saying our paper was valueless: we didn’t just speculate, we provided careful data analysis. The thing is, though, that the pattern we found, that big swings in Obama support could mostly be explained by differential nonresponse, surprised us. It wasn’t what we expected, it’s not something we thought about at all in our 1993 paper, and it took us awhile to digest this finding. But Palko, Silver, and Rivers had already laid out the whole story, all the way including the feedback mechanism by which small swings in vote preference are magnified into big swings in the polls, with all this connecting to the rise in survey nonresponse.

P.S. I wrote the post awhile ago. In the meantime, I gave that talk, “Wrong Again! 30+ Years of Statistical Mistakes,” for Jared’s New York R conference. Here’s the link to the talk.