There is an obvious aspect to this that is often overlooking in
attempts to understand and predict voter behavior, which is that
human beings are social animals and act not just for their own
welfare, but for the welfare of the society. To understand the
behavior of social animals, one must look not just from the
perspective of a society's welfare being some kind of sum of
individual welfare values, but also as if the society is itself a
kind of decision-making organism, making decisions on behalf of the
society as a whole.
"Society" here can range from a small collection of individuals all
the way up to every member of a species, and maybe even more than
that. In analyzing elections, we may think of collections of voters
who act coherently, and the application of game theory as if there is
no issue of "common welfare," of, say, voting as a ritual in which a
collective good is developed (or attempted), is going to be impoverished.
As has been pointed out, the individual expected value of voting is
miniscule, being the probability that my vote will affect the
outcome, times the value of that outcome over the other
possibilities, which usually is only one (in partisan elections in
the U.S., I have in mind).
But I don't vote because of that expected value; rather I behave as
part of a whole class of voters who, in some way or other, think like
me. Voting is a ritual in which we engage for the value that the
participation itself gives us, a sense of satisfaction, perhaps, at
having fulfilled a civic duty, and, collectively, of having exercised
the power we have instead of taking no responsibility for it.
Different voters may have different motivations, and many may never
even give the "expected value of voting" a second thought. They know
instinctively that by voting they are participating in something
larger than themselves. At that moment, "we" are making a decision,
or trying to.
----
Election-Methods mailing list - see http://electorama.com/em for list info
