Mike wrote: >So far as I'm aware, the results-reporting in Australia's IRV elections >isn't of a type that would reveal the problems that we've discussed. >All I've heard of is reporting of how many votes each candidate had >in each round, with nothing about the rankings or pairwise comparisons.
I poked around on the web site of the Australian Parliament (www.aph.gov.au). The House of Representatives (elected from single-member constituencies with IRV) has 2 main parties, a handful of independents, and a handful of people from one other party. The Senate, elected by PR, has 7 parties and a couple independents (although some parties only have one or two members). I don't know the PR method used. This more or less persuades me that IRV will not be as fruitful as one might hope. It's an open question how good Approval is in practice. I understand that Russia uses it for single-member elections, but half of their version of the House of Representatives (Duma? can't recall) is elected by PR. Although I am quite confident of the theoretical advantages of Approval, as a scientist I reserve final judgement until I see the data. Alex
