“Why Does Canvassing Work?”
those contributions to the ACLU or Greenpeace will alter any of one of President Bush’s
policies relating to the “War on Terror” or the environment? Even if one believed that
the benefit B of such a policy change was enormous, p would remain prohibitively low by
any reasonable estimate, thereby blocking the donation from transpiring. How can we
explain the thousands of who respond to canvassers by contributing or going out to polls,
sometimes weeks after the canvasser has gone?
There are of course more than a few answers to this question. One recent review
suggests sorting alternatives to Downs’ instrumental voting model into seven groups: (1)
consumption benefits of voting models, under which he includes both “civic duty” and
“expressive voting” formulas, (2) ethical (altruistic) voter models, (3) minimax regret
models, (4) game theoretic approaches, (5) group-based models, (6) information models,
under which he includes “bounded rationality” and (7) learning models (Geys 2006).
Expressive rationality and theories of altruistic individuals are the strongest contenders
because both offer compelling reasons why an individual might act against their own
narrow self-interest even when separated from their normal.
The altruistic theory offers
a powerful explanation of canvassing by accounting for a link individual level attributes
that would predispose one to caring about politics and listening to strangers (including
altruism, trust and optimism). Expressive rationality can tell us more about individual
motivation, canvass dynamics, campaign momentum and the ways in which identity may
2
What one recent field experiment has termed the neighborhood “surveillance effect” can offer at best a
partial explanation of canvassing because at the very least it cannot explain the success of street fundraising
in outdoor urban areas where donors are extremely unlikely to know one another (Gerber et. al 2008; cf.
Knack and Kropf 1998). The ongoing success of canvassing, both fundraising and electoral, is consistent
with “NICENESS” strategies (Bendor et. al. 1991, 1997), but canvassing cannot be well explained in terms
of models which are premised on repeated interaction between players because of the unlikelihood that the
canvasser and the individuals they speak to will ever see each other again (c.f. Bufacchi 2001). One might
ask about the social pressure the person canvassed feels from the canvasser, however this question does not
answer who feels pressure or why (e.g. is it their sense of altruism or something to do with their expressive
identity) or why exactly the canvassed person might feel better responding affirmatively. Further, the
possibility that some feel pressured does not rule out the possibility that others don’t.
Research Agenda
Mohanty 3
Convention