[EM] Student government - what voting system to recommend?

[Abd ul-Rahman Lomax]

At 10:40 AM 4/23/2007, Howard Swerdfeger wrote:
> > Range is expressive and it is able to treat these two different types
> > of "Pro Wrestlers" differently. Its problem is that it in practice
> > easily becomes Approval (only min and max values used) in competitive
> > elections.
>
>does it?
>I have seen arguments stating that a knowledgeable voter would alter
>there preferences in this manner. But I am unsure if this would happen
>in the reality of a large scale (>10^5 voters) election.

Yes, it is often asserted as a "criticism" of Range that it "easily 
becomes Approval." This is often asserted as if it were a known 
thing, which is what we see in what Mr. Swerdfeger quoted.

The error is in assuming that voters are, as a whole, dedicated 
supporters of single candidates. *Parties* may encourage 
bullet-voting in close elections, but they would be foolish to, for 
example, try to get the supporters of minor third parties, with no 
chance of winning the election, to bullet-vote. They could actually 
lower their own vote by doing this!

The analysis that leads to a conclusion that voters will bullet-vote 
under Range is based on an idea that voters prefer getting their 
favorite over even the smallest risk that their vote will elect 
another candidate *who is acceptable to them*. I'd suggest that this 
is not rational behavior for voters, and that, while some will indeed 
behave that way, many others will not.

Voting Approval style in Range is a simplified way of voting that 
many will doubtless follow. It is not harmful, so that this may 
happen to various degrees is not a criticism of Range. If it happens 
that, consistently, there is so little use of intermediate ratings 
that they do not affect outcomes, then it would be perfectly 
reasonable for the decision to be made to drop Range. For Approval, 
which is totally easy on the ballot and in the counting. Range isn't 
bad, though, and I suspect that, instead, the decision would be to 
drop the Range resolution to, perhaps, Range 3, with -1, 0, +1 as the 
votes. (This sets up midrange as the default vote for blanks.)