Scholars have long been concerned with voter turnout (Gosnell 1927). Interest further
increased with the consistent decline in turnout after 1960 (ending in 1992) or, if one accepts the
arguments of McDonald and Popkin (2001), with the failure of turnout to increase despite
increasing levels of education and the efforts of many states to make registration and voting
easier (culminating in the Motor Voter legislation). There have been three main lines of
questioning.
One, following Gosnell, has focused on how message modes, such as emotional versus
rational messages (Hartman 1936), affect the vote and how they interact with individual
personality characteristics like authoritarianism (Lavine et al. 1999), self-monitoring (Lavine and
Snyder 1996), or identification with a major party (Gerber and Green 2000a). The second
examines the effects of different means of communicating messages, such as personal
canvassing, direct mail and telephone calls (Gerber and Green 2000b). The third assesses the
impact of mobilization and recruitment efforts by parties or through social networks such as
churches (Gerber, Green, and Green 2003; Rosenstone and Hansen 1993).
Gerber and Green (and their co-authors) have taken the field several steps forward with a
series of field experiments that primarily explore the latter two questions. This methodology has
allowed them to draw causal conclusions about the impact and cost effectiveness of different
kinds of communication and the comparative efficacy of partisan and non-partisan get-out-the-
vote (GOTV) drives (Green and Gerber 2004). In this paper we continue the examination of the
impact of partisan and non-partisan mobilization efforts, as well as of different methods of
communication. Our aims are somewhat different, however. Our concern is less with the
effectiveness of the efforts than with the question of how these efforts are targeted, both by non-
partisan groups and with regard to the efficiency with which the major parties appear to identify
who they should mobilize. Scholars know that certain kinds of mobilization work, and that
2
Convention