Do yard signs win votes?
The short answer
- Lawn signs seem to win a few votes: combining four U.S. experiments in a way not planned in advance, precincts picked at random to get signs gave the advertised side an estimated 1.7 percentage points more of the vote. (Green et al., 2016)
- That gain is probably real but small, and unlikely to change the winner of a race that would otherwise be decided by more than a few percentage points. (Green et al., 2016)
- In those experiments, signs had essentially no effect on how many people voted, so the authors read any gain as coming mainly from changing some voters' choices. (Green et al., 2016)
What the studies found
- The four experiments covered a 2012 U.S. House race in New York, a 2013 Democratic primary for mayor of Albany, the 2013 Virginia governor's race and a 2015 Republican primary for county commissioner in Pennsylvania. (Green et al., 2016, online appendix)
- Researchers put signs in randomly chosen voting precincts; an estimate combining the four races, not planned in advance, finds precincts assigned signs gave the advertised side 1.7 percentage points more of the vote than precincts with no signs nearby. (Green et al., 2016)
- The authors conclude the effect of signs is probably greater than zero but unlikely to be big enough to change a race that would otherwise be decided by more than a few percentage points. (Green et al., 2016)
- Taken one at a time, each experiment gives unclear evidence, because each had too few precincts to measure a small effect with confidence; the finding comes from pooling all four. (Green et al., 2016)
- In a Virginia governor's race, a group's signs attacking candidate Terry McAuliffe raised his Republican opponent Ken Cuccinelli's vote share by 1.8 percentage points in precincts that got them, a result unlikely to be chance. (Green et al., 2016)
- The only test of signs in supporters' own yards, a Democratic primary for mayor of Albany, New York, gave results too uncertain to show whether the signs helped or hurt on average. (Green et al., 2016)
- There is some evidence that signs also helped in neighboring precincts that got none: an estimate combining the four races, not planned in advance, puts the gain at 1.5 percentage points. (Green et al., 2016)
- Lawn signs had essentially no effect on how many people voted, so the authors conclude any gain in vote share came mainly from persuasion, not from bringing out more voters. (Green et al., 2016)
- Counting only precincts that got signs, the authors estimate the signs cost about $3.18 per extra vote, with a likely range from $1.70 to $13.71. (Green et al., 2016)
- Lab experiments find that name recognition, simply knowing a candidate's name, can affect support for that candidate, and a key reason is that voters infer a familiar name can win. (Kam and Zechmeister, 2013)
- The same authors report a field study using yard signs during a county election, which they say shows the name-recognition effect holds outside the lab. (Kam and Zechmeister, 2013)
- A study of the 2008 U.S. presidential race finds that yard signs spread not only from campaigns down, but through voters' own initiative and their social networks. (Makse and Sokhey, 2014)
What the evidence can't tell us
- The pull of a familiar name fades when voters have a more useful cue, such as knowing which candidate is the incumbent, the person already in office. (Kam and Zechmeister, 2013)
- The sign experiments cover only four races in three states, and their authors say it remains to be seen whether the results hold elsewhere, calling for much larger experiments. (Green et al., 2016)
- The authors say more experiments are needed to learn why signs work; for the remaining idea, that signs show a candidate can win, the evidence is unclear because all four sign campaigns could have sent that signal. (Green et al., 2016)
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.
- Donald P. Green, Jonathan S. Krasno, Alexander Coppock, Benjamin D. Farrer, Brandon Lenoir and Joshua N. Zingher (2016). The effects of lawn signs on vote outcomes: Results from four randomized field experiments. Electoral Studies 41; read as the published version posted by co-author Alexander Coppock on his own site. Free copy · Publisher's page
- Donald P. Green, Jonathan S. Krasno, Alexander Coppock, Benjamin Farrer, Brandon Lenoir and Josh Zingher (2016). Online Appendix for: The Effects of Lawn Signs on Vote Outcomes: Results from Four Randomized Field Experiments (dated December 28, 2015). Online appendix to Electoral Studies 41, posted by co-author Alexander Coppock on his own site. Free copy
- Cindy D. Kam and Elizabeth J. Zechmeister (2013). Name Recognition and Candidate Support. American Journal of Political Science 57(4). Free copy · Publisher's page (summary only: no figures taken from it)
- Todd Makse and Anand E. Sokhey (2014). The Displaying of Yard Signs as a Form of Political Participation. Political Behavior 36(1); abstract read on the Florida International University research record. Free copy · Publisher's page (summary only: no figures taken from it)