At 03:28 PM 8/27/2006, Michael Poole wrote: >There are well-known cases where exit polls get the margins wrong >(e.g. Phillipines 2004) or even the results wrong (e.g. USA 2004).
I'd hardly call USA 2004 a "well-known case" of this. As Mr. Suter pointed out, serious controversy still exists regarding that situation. In the absence of some kind of verification, we really can't know for sure which occurred: error due to faulty polling, error due to actual deviation between voter actions and voter poll answers, or error due to vote fraud or other tabulation problems. > It >is extremely hard to correct for sampling error in exit polling, >making the overall polling accuracy comparable to certain alternatives >such as phone polls. For electoral results, both of those are >significantly more accurate than, for example, random polling in other >public places. I think that exit polling is much more likely to be accurate than phone polling, because it polls actual voters, reasonably well-verified as having voted. Sure, there are still sources of error; however, it should be possible to study the correlation between exit polls and actual election results. Serious anomalies, as in USA 2004, especially if only occurring in specific locations, could be grounds for further investigation. It has always seemed incorrect to me that the only persons having standing to force recounts are candidates. Don't the voters have a right to know what actually happened? If vote fraud happened, that the near-winner concedes does not make the fraud moot. Not only do we have a need and right to know, but if there is a sign of fraud, it should be investigated, because such fraud is generally illegal. (There are kinds of fraud that are not illegal, though perhaps they ought to be. Without charging that this actually happened, suppose a certain official, nominally a Democrat, decided to design a ballot which any expert could expect would skew the results. Nobody noticed and objected, and so this ballot was actually used and functioned as the plot intended. And thus the United States ended up with a President who not only lost the popular vote by a considerable margin, but only narrowly won the electoral vote because of the entire set of votes from Florida being awarded to that candidate due to a few-hundred vote margin, which itself was nearly certain, from later investigation, to be incorrect, and which itself depended on many little "nudges," each not specifically illegal but possibly illegal in the context of a conspiracy. Tossing alleged felons, based on records provided from other states, was not illegal, and that it just happened to eliminate, in error, far more Democratic voters than Republicans, well, that's the breaks. Get over it.) Exit polling to make an educated estimate of the effect of different voting methods is an excellent idea. It is non-binding, it would only provide information, but information that is not possible to obtain with such accuracy in any other way, I'd suggest. Would improved election methods shift outcomes? Most of us are quite certain that it would; simply allowing overvotes, by striking the line in the election code that prohibits counting them, would have moved US Presidential Florida much more toward the realization of the intention of the majority of voters. It is quite likely, I'd suggest, that the Approval Voting that would result from this simple change would shift *many* election results. This could partly be studied through analysis of actual ballots; but we can't tell from that how voters would shift their votes once they understood that overvotes would be allowed. Certainly many would. I might even consider voting for a Green! or, in some races, for that matter, a Libertarian. However, personally I'd put the effort into forging the real missing link: direct organization of the public outside of government, and, indeed, that is what I'm about. If the public were organized in the libertarian/anarchist method we call FA/DP, which is institutionally unbiased (i.e., it is libertarian, not Libertarian, direct/representative democracy, not Democratic, deliberative, not aggregative), we wouldn't have to worry about election fraud, for we would know, in advance of the election, the certain result, at least in most cases. And, as with the Orange Revolution in the Ukraine, if the oligarchs rigged the election, we simply would not accept it. If we put 10% of the effort we put into trying to manage the content of the political system into studying and reforming the *structure*, we'd find content much easier to manage. But, of course, strictly speaking, we are not advocating any change in the legal structure, only something new, something added to it: direct organization of the people, which has never before been attempted on the scale of a nation-state except through coercive or at least nondemocratic, top-down structures, which, like all such structures, if successful, continue to act to preserve themselves and the special interests of the oligarchs at the top. There is a better way, and it may be that the time has come for it. At least we are trying. FA/DP, coming soon to a party near you. ---- election-methods mailing list - see http://electorama.com/em for list info
