Robo-Polls: Taking Cues from Traditional Sources?

Abstract

After the 2012 Republican New Hampshire primary, there were 159 poll results released prior to the subsequent nomination contests. More than two-thirds of these polls relied on “Interactive Voice Recognition” (IVR) software. In this research note, we evaluate the ability of polls to predict the vote-share for the Republican candidates Romney, Santorum and Gingrich. We find no difference in the average accuracy of IVR and traditional human polls, but IVR polls conducted prior to human polls do significantly worse than traditional human polls even after controlling for characteristics of the states, polls, and electoral environment. These findings provide suggestive evidence that pollsters may take cues from one another given the stakes involved. If so, reported polls should not be assumed to be independent of one another and so-called “poll-of-polls” will therefore be misleadingly precise.

Keywords: Elections, Polling