How social pressure increases voter turnout: Evidence from a large-scale field experiment
New research by political scientists concludes that direct mail campaigns which include a social pressure aspect are more effective at increasing voter turnout and are cheaper than other forms of voter mobilization, including door-to-door or telephone canvassing.
Conducted by political scientists Alan S. Gerber (Yale University), Donald P. Green (Yale University), and Christopher W. Larimer (University of Northern Iowa), these findings are presented in an article entitled “Social Pressure and Voter Turnout: Evidence from a Large-Scale Field Experiment.” The complete article appears in the February issue of the American Political Science Review, a journal of the American Political Science Association (APSA), and is available online at http://www.apsanet.org/imgtest/APSRFeb08Gerberetal.pdf
Prior to the August 2006 primary election in Michigan, the researchers sent out one of four various mailings to 80,000 households encouraging them to vote—with gradually increasing levels of social pressure. The first mailing reminded voters that voting is a civic duty. The second mailing informed the voters that researchers would study their turnout based on public records. The third mailing listed a record of voter turnout among those in the household. The fourth mailing displayed both the neighborhood and household voter turnout. The third and fourth mailings also suggested that there would be a follow-up letter after the upcoming election, reporting on their household or neighborhood voter turnout.
The authors found that when it comes to voting, people are more likely to conform to powerful social norms—like viewing of voting as a civic duty—if they expect that their behavior will be made public. For example, after households were shown their own voting record, their turnout rose to 34.5%, a 4.5% increase over the control group’s voting rate of 29.7%. “Even more dramatic is the effect of showing households their own voter record and the voting records of their neighbors,” note the authors. Voter turnout among households exposed to this method was 37.8%, an increase of 8.1% over the control group.
This remarkable increase in turnout, observe the authors, “exceeds the effect of live phone calls and rivals the effect of face-to face contact with canvassers conducting get-out-the vote campaigns.” By comparison, policy interventions such as Election Day registration or vote-by-mail, which are widely debated today and seek to increase turnout by lowering the costs of voting, are thought to have effects of 3% or less. Moreover, in terms of sheer cost efficiency, mailings that exert social pressure cost between $1.93-$3.24 per vote, far outperforming the roughly $20 per vote for door-to-door canvassing or $35 per vote for phone banks.
Source: American Political Science Association