Do candidate visits change votes?
The short answer
- Candidate visits seem to move few votes: in the closing stage of the 2012 presidential race, one study finds visits had only a small effect on how people planned to vote, and it faded fast. (Wood, 2016)
- A visit can stir up both sides: in a 2006 Texas experiment, the governor's appearances raised support for him and also for his Democratic opponent. (Shaw and Gimpel, 2012)
What the studies found
- A survey in the last stage of the 2012 U.S. presidential race finds that people knew little about candidate visits, and that visits had only a small, short-lived effect on how people planned to vote. (Wood, 2016)
- The same study finds that state-level politics shapes where presidential candidates go, and reports that the visits themselves have little effect on local media markets. (Wood, 2016)
- When Texas Governor Rick Perry's January 2006 appearances were assigned to media markets at random, support rose for him and for his Democratic opponent, and the effects did not fade as fast as some studies find for TV ads. (Shaw and Gimpel, 2012)
- In that Texas experiment, the governor's appearances more clearly brought in campaign donations and volunteers than votes. (Shaw and Gimpel, 2012)
- In swing-state media markets in 2016, the share of survey takers planning to vote for Donald Trump rose 4.5 percentage points in the 10 days after a Trump rally, compared with swing-state markets without one. (Snyder and Yousaf, 2020)
- That effect lasted about one week, and the same study finds no consistent effect from rallies by Hillary Clinton in 2016, Barack Obama and Mitt Romney in 2012, or Obama and John McCain in 2008. (Snyder and Yousaf, 2020)
- In 2016 the presidential candidates went mostly to swing states: from September 22 to Election Day, all but 2 of Trump's 76 rallies and all but 2 of Clinton's 58 were in swing states. (Snyder and Yousaf, 2020)
- In swing states in 2016, a Trump visit raised donations to Clinton's campaign in that media market by 7 percent that day, more than it raised donations to his own campaign. (Heersink, Peterson and Peterson, 2021)
What the evidence can't tell us
- The donation study measures giving on the day of a visit, not votes, and its authors say the effects are too small to change the result of the election. (Heersink, Peterson and Peterson, 2021)
- The rally study mostly measures survey answers; its look at 2016 vote counts links final-week Trump rallies to more Republican votes and turnout, but places and timing were not random, and an unreported check found little effect on vote share. (Snyder and Yousaf, 2020)
- The Texas experiment tested appearances by a single governor during a few days in January 2006. (Shaw and Gimpel, 2012)
Sources
Every study here was read in a free copy, linked below. A number on this page comes only from a study read in full; one read only in its summary is described in words.
- Thomas Wood (2016). What The Heck Are We Doing in Ottumwa, Anyway? Presidential Candidate Visits and Their Political Consequence. The ANNALS of the American Academy of Political and Social Science 667(1). Free copy · Publisher's page (summary only: no figures taken from it)
- Daron R. Shaw and James G. Gimpel (2012). What if We Randomize the Governor's Schedule? Evidence on Campaign Appearance Effects From a Texas Field Experiment. Political Communication 29(2); abstract read on the ResearchGate page of the copy uploaded by James G. Gimpel. Free copy · Publisher's page (summary only: no figures taken from it)
- James M. Snyder Jr. and Hasin Yousaf (2020). Making Rallies Great Again: The Effects of Presidential Campaign Rallies on Voter Behavior, 2008-2016. NBER Working Paper 28043, October 2020 (a working paper, not peer reviewed). Free copy · Publisher's page
- Boris Heersink, Brenton D. Peterson and Jordan Carr Peterson (2021). Mobilization and Countermobilization: The Effect of Candidate Visits on Campaign Donations in the 2016 Presidential Election. Journal of Politics 83(4); read as the authors' posted version of May 16, 2020. Free copy · Publisher's page