Does knocking on doors win votes, or just bring out voters?
The short answer
- Knocking on doors brings out voters: across U.S. experiments, a visit that reaches someone raises the chance they vote by about 2.5 percentage points on average. (Bhatti, Dahlgaard, Hansen and Hansen, 2019)
- Changing minds is much harder: in U.S. general elections, the authors' best guess is that campaign contact in the last two months, door-knocking included, persuades about 1 in 800 voters, which they call substantively zero. (Kalla and Broockman, 2018)
- Persuasion does show up in ballot measure votes, where voters cannot lean on party labels; in primaries (contests inside one party) the estimated effect is larger but too uncertain to be sure. (Kalla and Broockman, 2018)
What the studies found
- In a 1998 New Haven, Connecticut experiment, people picked for a nonpartisan (not for any party) canvassing visit voted 2.43 percentage points more than people not picked, though canvassers reached only 28% of them. (Gerber and Green, 2000)
- Counting only the people canvassers actually spoke with, the same New Haven visits raised the chance of voting by 8.7 percentage points. (Gerber and Green, 2000)
- In that experiment, face-to-face canvassing raised turnout a lot, mail raised it slightly, and phone calls not at all. (Gerber and Green, 2000)
- In Western Europe, pooled door-knocking experiments show a much smaller average rise of 0.78 percentage points among people reached, too uncertain to say whether canvassing works there at all. (Bhatti, Dahlgaard, Hansen and Hansen, 2019)
- Pooling U.S. studies, including the authors' own, door-knocking or phone calls in the last two months before a general election have an estimated average effect on vote choice of 0.58 percentage points, which cannot be told apart from zero. (Kalla and Broockman, 2018)
- That near-zero effect in general elections holds for races at every level of government, close or not. (Kalla and Broockman, 2018)
- Contact more than two months before a general election often seems to persuade when measured right away, but in the two studies that checked later, the effect faded. (Kalla and Broockman, 2018)
- In Philadelphia's 2015 Democratic primary for mayor, door-knocking six weeks out raised support for the canvassers' endorsed candidate by about 11 percentage points a week later. (Kalla and Broockman, 2018)
- Pooling U.S. door-knocking experiments reviewed up to 2013, one study finds that being reached by a canvasser raises a person's chance of voting by 2.50 percentage points on average. (Bhatti, Dahlgaard, Hansen and Hansen, 2019)
- Pooling earlier U.S. experiments on mail, calls, door-knocking and ads in the last two months before a general election, the authors' best guess is that contact persuades about 1 in 800 voters, and at most about 1 in 175. (Kalla and Broockman, 2018)
- Pooling U.S. experiments, campaign contact shifted ballot measure votes by an estimated 3.23 percentage points among people reached; in primaries the pooled estimate, 4.51 percentage points, was larger but too uncertain to rule out zero. (Kalla and Broockman, 2018)
What the evidence can't tell us
- A door-knocking effect can be reported per person a campaign tries to reach or per person actually reached; in New Haven in 1998 canvassers reached only 28% of the people assigned, so the second figure was much larger. (Gerber and Green, 2000)
- The persuasion experiments do not measure the effects of a candidate's own qualities or positions, or the best split of campaign money between turnout and persuasion. (Kalla and Broockman, 2018)
- Most door-knocking experiments come from the United States; the European studies are few, all from Western Europe, and many are small. (Bhatti, Dahlgaard, Hansen and Hansen, 2019)
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.
- Alan S. Gerber and Donald P. Green (2000). The Effects of Canvassing, Telephone Calls, and Direct Mail on Voter Turnout: A Field Experiment. American Political Science Review 94(3). Free copy · Publisher's page
- Yosef Bhatti, Jens Olav Dahlgaard, Jonas Hedegaard Hansen and Kasper M. Hansen (2019). Is Door-to-Door Canvassing Effective in Europe? Evidence from a Meta-study across Six European Countries. British Journal of Political Science 49 (first published online 2016). Free copy · Publisher's page
- Joshua L. Kalla and David E. Broockman (2018). The Minimal Persuasive Effects of Campaign Contact in General Elections: Evidence from 49 Field Experiments. American Political Science Review 112(1); read as the authors' accepted manuscript of July 26, 2017 on eScholarship. Free copy · Publisher's page