<HTML><FONT FACE=arial,helvetica><HTML><FONT SIZE=2 PTSIZE=10 FAMILY="SANSSERIF" FACE="Arial" LANG="0">Bart Ingles wrote:<BR>
<BR>
>There is no way to accurately determine approval voting results using<BR>
>this input format. About the best you can do is to assume that for the<BR>
>A>B voters, half approve both A and B, and the other half approve only<BR>
>A. This would make approval voting equivalent to Borda, at least where<BR>
>fully ranked ballots are concerned. <BR>
<BR>
The model assumes that voters either use a good Approval strategy (Rob Le Grand's strategy A) or approve all candidates they rank. The proportion of voters using strategic voting and non-strategic voting can be varied.<BR>
<BR>
>Also, I notice that you allow equal last-choice preferences (e.g.<BR>
>A>B=C), but not equal first preferences (A=B>C). Doing one but not the<BR>
>other would bias the results. But then the only reasonable way to<BR>
>handle (A=B>C) for IRV or Plurality would be to assume that half vote<BR>
>one way, and half the other.<BR>
<BR>
This is not correct. The model allows truncation but not equal preference. A>B means voters rank A first and B second and C not at all.In strategic voting whether A>B voters approve A or A and B depends on how the voters respond to the information in an opinion poll using strategy A. In non-strategic voting the voters approve all candidates they like and hence A>B voters approve both A and B.<BR>
<BR>
Approval was also problematic to model.<BR>
<BR>
David Gamble<BR>
<BR>
<BR>
<BR>
<BR>
<BR>
</FONT></HTML>