PEP 308: Alternative conditional operator forms

[Laura Creighton]

> With the yes/no voters, you hope that it's a good cross-section of the
> community, but of course there's no way to be sure of that.  But
> certainly someone strongly in favor of the proposal is just as likely to
> vote as someone strongly against it, so one would expect that as long as
> the vote ballot reaches a wide enough audience, the yes/no voters should
> be in the right proportions.  But you _know_ for a fact that the
> abstentions will not be.

Anybody who believes that yes/no voters show up at the poll in the
same proportion as the yes/no ers in the general population, needs to
read up on convenience sampling.  We know lots about voting now, and
one thing that is clear is that the subject matter of the vote is
highly relevant.  In general, people who want change are more
motivated to vote than people who want things to stay as they are.
They are also more likley to be paying attention to the fact that
change is being voted.  This is why you want sunny weather on voting
day when your political party is trying to be re-elected, and why you
pray for rain if you want to vote those incumbent bums out.  This is
not the case, however, when the proposed change is a new tax.  Then
people who dislike paying taxes come out in droves to try to sink the
proposed measure.  This is why if you want something that costs money,
you campaign on how great it is going to be, and try to only worry
about the costs after the election.

I have no idea whether this proposal is more likely to motivate those
who sincerely want to vote yes than those who sincerely want to vote
no, or vice versa.  But I have perfect faith that if the proportions
we get somehow do accurately match the proportions of the population
at large (something we can never know) then this will only be by
coincidence.

Laura