Hi folks, On the approval vs. IRV question, I used to lean more towards IRV, but these days I've been leaning more in the other direction. It occurred to me today that the re-election of George Bush may be partly responsible for this. First, I don't think that GB should have been re-elected, in that the anybody but Bush vote was so large and passionate. Furthermore, his own support was not all that solid, and his victory relied very heavily on portraying the other candidate as unacceptable, a tactic that would have much more limited effectiveness if he was facing more than one viable contender. I think that the strength of the anybody but Bush faction would have been fatal to his campaign in an approval election, because voters would be unusually willing to approve compromise candidates whom they preferred to Bush. If IRV had been used in 2004 however, it's harder for me to imagine Bush losing. Most likely he would have faced Kerry in the last round, and given that, the result would not have been much different. Unlike 2000, votes for third party candidates were not enough to fill the margin between the top two, so I think that using IRV instead of plurality would not have had a major impact. As long as the Bush campaign was able to predict whom he would face in the last round, they would have been able to effectively target that candidate for attack. I'm not saying that Kerry would have won under approval. I seem to remember reading a poll in the fall of 2004 in which Edwards had the highest approval score out of the four P/VP candidates. Perhaps Edwards or someone similar would have won, e.g. Wes Clark or John McCain. I'm not sure if Kerry was a polar candidate from the beginning (I feel fairly sure that he didn't set out to be one), but the pressures of a plurality election (i.e. the strong push toward negative campaigning) may have made him into one by the end. In an approval election, I imagine that negative campaigning wouldn't be quite as rampant. Anyway, that's my story. I'm not intending to make a partisan point here, but rather I'm saying that the election seems to illustrate a general type of scenario in which approval may be more likely than IRV to choose somewhat less-polar candidates. Of course, this is just anecdote, not really an analytical argument, but I thought that others might find it slightly interesting. It may help to explain why I now feel that approval voting is worth pursuing. my best, James
---- Election-methods mailing list - see http://electorama.com/em for list info