<HTML><FONT FACE=arial,helvetica><HTML><FONT SIZE=2 PTSIZE=10 FAMILY="SANSSERIF" FACE="Arial" LANG="0">A couple of points:<BR>
<BR>
Bart Ingles wrote:<BR>
<BR>
>Dgamble997@aol.com wrote:<BR>
>> <BR>
>> Bart Ingles wrote:<BR>
>> <BR>
>> >But truncation is equivalent to equal last-choice preference for all of<BR>
>> >the methods listed below.<BR>
>> <BR>
>> Yes, it is equivalent but expressing an equal preference for two or<BR>
>> more candidates is generally considered as different to not ranking<BR>
>> candidates you don't like.<BR>
<BR>
>I don't understand this. How can it be both equivalent and different? <BR>
>I guess I don't understand the distinction you are making here<BR>
<BR>
Giving two candidates you feel are worth ranking an equal ranking is a positive expression that you a/ like them both and b/ like them to the same extent. Not ranking them can mean that you prefer one to the other but consider neither worthy of support.<BR>
<BR>
For example on the ballot paper there is a right wing conservative and a fascist. If I had to rank all candidates I'd rank the right wing conservative above the fascist. If I had the option to truncate I'd support neither of them because I like neither of them though one is preferable to the other ( both are extremely low utility).<BR>
<BR>
>In a realistic scenario, it's inconceivable that individual voters (even<BR>
>hypothetical ones) would know their own preference orders, yet fail to<BR>
>know their own utility levels for each candidate.<BR>
<BR>
I disagree, you have say 4 flavours of ice cream ( chocolate, mint, strawberry and vanilla). What is the easier task to establish a ranking or give each a cardinal utility rating? I finding ranking the ice creams easier than giving them a utility rating on a say a scale of 0 to 100.<BR>
<BR>
>Since we don't know how the voters would rate the candidates, but have<BR>
>to assume the voters themselves know, it seems reasonable to assume a<BR>
>range of opinions with half the voters rating their middle candidate<BR>
>above average, and half below average. Thus in the zero-info case, I<BR>
>would expect to see half the voters approve exactly one candidate, and<BR>
>the other half to approve exactly two candidates. It turns out that<BR>
>this gives results identical to Borda.<BR>
<BR>
Generally what do people think is more realistic in translating the ranked ballot A>B into an Approval vote in a zero information case with 3 candidates? <BR>
<BR>
1/ That the voters approve everybody they rank.<BR>
<BR>
2/ That half the A>B voters approve only A and half approve A and B.<BR>
<BR>
3/ Something else.<BR>
<BR>
What about irrational/non-strategic voters?<BR>
<BR>
In the Yahoo files there is a paper by Steven Brams: Going from Theory to Practice: The Mixed Success of Approval Voting. In the elections detailed a small number of voters use obviously poor strategy -approving all candidates or protest strategy approving none of the candidates - both of which have the same effect on determining the winner as not voting. I wonder how many other people in those elections weren't voting strategically/rationality but we can't detect this because the votes they cast appear to make sense though they didn't come to that decision by a rational/strategic process ( an awful question that probably can't be answered).<BR>
<BR>
David Gamble<BR>
<BR>
David Gamble<BR>
<BR>
<BR>
<BR>
</FONT></HTML>