How Different Are Telephoning and Canvassing? Results from a ‘Get Out the Vote’ Field Experiment in the British 2005 General Election
Can the positive impact of non-partisan ‘Get Out the Vote’ (GOTV) campaigns be generalized to a variety of institutional and cultural contexts? Gerber, Green and colleagues tested for the effects of these campaigns in a series of pioneering field experiments, which show that a face-to-face contact from a non-partisan source, carried out by a field force calling at the homes of citizens seeking to persuade them to vote, can increase voter turnout. Further experiments find that telephoning has an impact ranging from ineffective to positive, depending on the nature of the call; and there are positive, if weaker, results for other forms of intervention, such as door postings and leafleting; none for e-mail; and weakly positive or null impacts from rote telephoning. Many of these results derive from single cases or from a limited number of research sites; however, the culmination of these findings allows political scientists to be confident of the impact and hierarchy of these interventions. Although GOTV studies of this kind cannot adjudicate authoritatively on theories of mobilization, the difference in impact between the types of intervention, in particular the greater success of personalized messages, implies that it is the personal and face-to-face basis of influence that has an effect, rather than the types of message received and the simple provision of information.
So far most of this kind of research has been carried out in the United States, which means that, even with its variety of groups and locations, the range of variation in the institutional frameworks and social conditions is limited to the one-country case.(Published Online May 13 2008)
a The authors wish to thank the University of Manchester and the McDougall Trust, which funded the project. They also wish to thank the research students who carried out the canvassing and Vision 22 for the telephone survey. The comments and support of colleagues at Manchester and from Don Green at Yale have been greatly appreciated. Helpful points were made at a panel on the experimental method in political science at the Annual Meeting of the American Political Science Association, Philadelphia, 2006.