Slashdot Mirror


Mathematicians: Elections Flawed

Nader-licious writes "Science News Online reports: 'With recent reports of malfunctioning voter machines and uncounted votes during primaries in Florida, Maryland, and elsewhere, reformers are once again clamoring for extensive changes. But while attention is focused on these familiar irregularities, a much more serious problem is being neglected: the fundamental flaws of the voting procedure itself. Mathematics are shedding light on questions about how well different voting procedures capture the will of the voters.' The verdict: the U.S. system might be the worst of the lot."

15 of 551 comments (clear)

  1. My view on "instant runoff" by lpontiac · · Score: 5, Interesting

    I'm an Australian, and we use the "instant runoff" system described in the article. My view on it is that rather than putting the most popular candidate into office, it keeps the least popular candidates out of office.

    There is a problem that the article neglected to mention - "how to vote" cards. Each candidate will generally recommend how they think people should vote - themselves first, naturally. The same sheep mentality that leads to 70% of the population voting for the same party every election leads to many religiously following these how to vote cards.

    The end result is a heap of wheeling and dealing between candidates for these "directed preferences." It even becomes a stick in between elections that the minor parties can use to beat a major party with; in a marginal seat, having a minor party favour you over your primary opposition can be the difference between winning and losing.

  2. The system in Australia by The+Original+Yama · · Score: 4, Interesting

    I'm quite partial to the Australian system (although I may be biased since I'm Australian). It is a variation on the simplistic British 'first-past-the-post' system. Basically, you number each candidate in the order you prefer, with #1 being your favourite candidate. When the votes are counted, they first tally all the #1 votes. If after that nobody has a clear majority (50% of the vote plus one), they count the #2 votes and add them to the #1 tallies. They keep doing that until someone gets a majority.

    What I like about this system is that you are not tying yourself to one candidate. Your vote won't be wasted if you vote #1 for a minority candidate, since if they don't win your next preferences may count. This also means that you're not necessarily guaranteed a win if you're in one of the larger parties.

    In the end (generally), you don't get an electorate that's split between people who did and didn't vote for the winner. Since everybody's preferences are taken into account, you get a decent compromise.

  3. The system won't change by rseuhs · · Score: 5, Interesting
    Currently, power is shared between Republicans and Democrats.

    Neither would be happy if the system would allow more than 2 parties to exist, so neither will ever agree to a substantial reform.

    1. Re:The system won't change by the+bluebrain · · Score: 5, Interesting
      • Currently, power is shared between Republicans and Democrats.
        Neither would be happy if the system would allow more than 2 parties to exist, so neither will ever agree to a substantial reform.
      Yup. That is what I was missing in the article, too. Obvious, really: voting theorists deal with the theory of voting, not with the mechanics of change, and the question of what is politically doable.
      Point being, not only the incumbents, but also the lower echelons grew up with the current system, and they know how to play it - play being exactly the right word. I expect few professional politicians would want to change the system, because the change would cause uncertainty - they would rather be big underdogs in the current system, than risk even the very small danger of becoming bit players in a new one.

      A thought on this issue: as you say, neither the Dems nor the Reps would like to see the rise of smaller parties, because this would erode their influence (power base), and they would even (gasp) co-operate to keep the system just the way it is. However, if a party smaller party does grow big enough to be an "annoyance" to the established ones, the one more hurt by the smaller party will bring the issues in this article up - hence this discussion.

      For instance: the Dems would be prepared to give the greens say 10 seats in the senate, if the Dems in turn get the presidency - quite a likely situation if the voting procedures are changed as described.

      The point to which I have been coming all the while is this: in a one-dimensional political spectrum, the "right" is stereotypically more disciplined than the "left", i.e., they have a stronger belief in law & order, hierarchical systems etc. This means that if there is a small rightist party (e.g. a bunch of neo-nazis by anything but name) in a situation similar to the one in Florida during the last presidential election, their adherents will be more likely to follow the "orders" of the party leadership to vote for Bush rather than their own candidate (because Bush is better than Gore, and their own candidate doesn't have a real chance to win) than their "leftist" counterparts.

      This means in turn that the results of the plurality vote in the US is not only the two-party system that we observe, but also a slight shift to the right.

      In closing I must say that in my opinion, replacing the plurality vote would be the single biggest step the US could undertake to enhance its political image, pretty much everywhere else in the world.
      --
      yes, we have no bananas
    2. Re:The system won't change by smileyy · · Score: 3, Interesting

      I am aware that the word "socialist" is part of the acronym "NSDAP" - put please don't be bamboozled into believing that that is what they actually were.

      That's probably similar to the trend that, any country that puts "Democratic" into its name, likeley isn't. For reference, the old German Democratic Republic.

      --
      pooptruck
    3. Re:The system won't change by salimma · · Score: 3, Interesting

      Indeed. Here in UK the Labour Party while in opposition (pre-1997) worked out an alliance with the Liberal Democrats, the third party in domestic politics, a central feature of said alliance is the introduction of some sort of proportional representation.

      This measure, which would benefit the Lib-Dems (consistently polling at around 20% but obtaining only 8% of the seats under the first-past-the-post system) was scuppered after Labour unexpectedly won a huge landslide in 1997 and kept a large part of its majority in the subsequent election in 2001. The carrot is still being dangled though..

      --
      Michel
      Fedora Project Contribut
  4. NOT flawed, designed not to capture will by NSParadox · · Score: 3, Interesting

    Political scientist have known for years that the US election system does not capture the "will" of the voter as well as a proportional representation system. This math is certainly not new. However, there are a number of drawbacks to other forms of election that should be expounded on.

    1: In proportional representation, there are more likely to be minority parties with elected officials who have extreme/radical viewpoints that are dissimilar to the viewpoints of the "average" voter. Because of the US' election system, no candidate can choose to isolate a significant portion of the population with his views and yet still be elected, to a large "smoothing out" extremist policy. While many feel that this is a bad thing, almost all extremist policy is not realistic to implement, and partial or full implementation of this policy can cause a good bit of damage.

    2: In proportional representation, the government is generally unitary in nature, meaning that the entire government is controlled by one party. Although there are more parties beyond the controlling party and another party represented, they still have a HUGE capability to control government policy. If the party in charge changes (and they often change), the entire government policy may change as well. Imagine if a country implemented social security, and then cancelled it 12 years later because the Socialists were replaced by Libertarians!

    3: Most other countries do not implement a form of federal government. While this may work for countries where there country is roughly the size of a US region, it makes interests associated with a geographical locale very difficult to achieve. While every vote should be equal (or as equal as possible), the reality is that interests are largely decided by the environment of the voter, and partitioning the environment, and tiering government, means more interests of more voters are going to be met without completely missing the interests of other voters.

    4: Most unitary governments do not have a strong set of checks and balances; i.e., judges and execute officials are appointed via the parliament/prime minister, and the prime minister is elected by the parliament. The effect of this election policy is similar again to point 2: a shift in political power can cause a dramatic shift in policy in a short period of time because there are fewer roadblocks between the will of the current parliament and the implementation of that will.

    Out of all of the election policies I've studied, IGNORING THE ELECTORAL COLLEGE (because it's a system with several undescribed states, if we were to somehow reach one of those states by having an election of an official "tied" in enough ways we wouldn't know what the next step would be), I prefer the US government system. It's not designed to reflect the will of the people right now. It's designed to reflect the long-term interests of the people after filtering out extreme views. Its perponderance of gridlock has prevented so many stupid things from happening it's totally uncountable. That being said, I like the way Australian government is structured, except I REALLY do not like the idea of being able to put multiple candidates on a list. Political scientist mathemeticians have shown that by being able to list multiple candidates on a piece of paper, it increases the voting power of a citizen to > 1, and they can use these voting lists to perform elaborate tricks to achieve an end result which might not effect the will of the voting populace at all.

    Tired of rambling, so I think I'll stop here.

    --
    Unless mankind redesigns itself .... robots will take over our world. (Stephen Hawking)
  5. We Don't Have A Federal Government... by theduck · · Score: 4, Interesting

    Most other countries do not implement a form of federal government.

    We (USA) don't have a federal government, at least not in the pure sense of the word. We have a national government.

    What's the difference? Whether power resides primarily on the state or national level. A federal government represents and is controlled by a federation of smaller political entities (states, in the USA) where the true power resides. A national government represents and is controlled by a single national political entity that might or might not be comprised of smaller political entities.

    The single best way to determine whether you have a federal or national government is where the primary power of taxation resides since a government can do nothing without revenue. The political struggle between the federalists and anti-federalists in the USA centered mainly around this point. Oddly enough, there was the same confusion between the terms "federal" and "national" back then. Apparently, the Federalist marketers got their mits onto that confusion first, because the Federalists were actually for a national government and the Anti-Federalists for a federal government.

    --
    How can we afford to ever sleep
    So sound again
    --ebtg
  6. Re:Absolutely wrong. by Kierthos · · Score: 3, Interesting

    Montana? Excessive influence? Sonny boy, with that statement, you've shown you know two things about the way the electoral college works: Jack and shit.

    Montana has three electors. Three. That's the least you can possibly have. There are 538 total electoral votes. You need 270 to "win" the Presidency.

    Unless it's a god-awful close election, Montana and other small states get ignored, because it's much more effective, politically, to focus on states like California (54 electoral votes), New York (33 electoral votes), Texas (32 electoral votes), and Florida (25 electoral votes).

    With those 4 states, you have 144 electoral votes. Just over half of the total you need. Throw in Ohio (21), Illinois (22), and Pennsylvania (23), and you're at 210 electoral votes. 60 shy of what you need, with 7 states.

    Get a few other states sewn up, and you're set.

    Montana and other rural states get screwed.

    The electoral college system should not be used as the end-all, be-all system. We need something that actually works.

    Kierthos

    --
    Mr. Hu is not a ninja.
  7. Re:Either Mod This Guy Up as Funny... by theduck · · Score: 3, Interesting

    Pardon me, but racist? Kindly explain how you got that from my comment!

    Anyway, your response indicates that you judge the success of a political system based on economic results. I must point out that the two are not necessarily related. However, since you're the original poster, I'll play on your playing field.

    Yes, China is currently growing very quickly. But keep in mind that it's much easier to achieve a high rate of growth when your productivity is lower (in this case, GDP/population).

    China is a totalitarian state (political system) that is beginning to realize that capitalism (economic system) works better than communism (economic system) in the real world. We have yet to see whether the economic freedom the Chinese government is beginning to allow will cause the populace to demand political freedom (some form of representative democracy; a family of political systems) as well.

    If the USA is considered to be a political-economic experiment at ~225 years, then the political-economic environment in China right now (~10 years old?) certainly qualifies as an early-stage experiment and any assessment of their ultimate success or failure should be considered to be, at best, a guess.

    Also note that I mentioned the Soviet Union in my reply to your post. I don't see you touting their growth rate.

    --
    How can we afford to ever sleep
    So sound again
    --ebtg
  8. Canada! Canada! by davedave · · Score: 3, Interesting

    Ah, you Americans would love the Canadian system. I've lost count of the number of political parties we have. Each constituency is pretty much guaranteed a Liberal, a Progressive Conservative, and an NDP candidate. And then, depending on if you're east or west of Ottawa, you'll have either a BQ candidate or a Reform party candidate. And then there are the various minor parties (The Green Party, the Natural Law party - Yogic Flying, it's all the rage! - the Pot Party, etc), and the independent candidates. So we're talking 6 + candidates per constituency, and it's a plurality vote. So, we end up with a party that had maybe 40% of the popular vote controlling 60 to 70% of the Seats in the House of Commons. Which is why we get a leader like Jean Chretien threatening to cram Kyoto down our throats without even consulting his First Ministers, and those damn grits are gonna NEP us Albertans all over again. And don't even get me started about the Senate!

    --
    A One that isn't cold is scarcely a One at all.
  9. French election turn out by autopr0n · · Score: 4, Interesting

    99% of voters were absolutely certain that the 2nd round would bring the good old traditional Center-Left vs Center-Right showdown (Jospin-Chirac in that case), so many people didn't even care to vote.

    Yet, france had a 70% voter turn out, Far higher then any US elections :P

    --
    autopr0n is like, down and stuff.
  10. One thing that stands out ... by tdelaney · · Score: 3, Interesting

    ... is that in both the instant runoff (as we have in Australia - we call it two-party preferred) and Borda systems, the final result *tends* to be either the first or second choice of the majority of voters. Of course, as the number of candidates increases, this obviously becomes less the case as votes are further split. But in any case, the final result will be from the upper half of the majority of voters preferences.

    As the concocted example shows, this is much less likely to occur in a plurality system. In fact, the plurality system actively works against this being the case where there are more than two candidates.

  11. Electoral College by Arandir · · Score: 3, Interesting

    The article says that the US system does not represent the will of the voters best because of the Electoral College. Duh! It's not supposed to!

    The Electoral College was set up to prevent the raw unfiltered will of the populace from ruling. It's purpose is to process and filter the will of the populace. This is a Good Thing(tm). The Electoral College is there for exactly the same reason that a President is being elected to begin with: the US political system is a representative republic, not a direct democracy.

    The whims of the poplulace changes daily. A look at pre-election polls over a period of a few weeks demonstrates this. The Electoral College helps filter these mood swings out.

    I realize that I am the last living person in the US who still likes the Electoral College, but that does not necessarily make me wrong.

    --
    A Government Is a Body of People, Usually Notably Ungoverned
  12. Generally good article, but.... by robla · · Score: 3, Interesting
    ...doesn't venture out much further than an article in Discover Magazine a couple years ago. It pits Brams against Saari, and says "you decide". This one, as opposed to the Discover article, talks about Instant Runoff more, though.

    The field is more complicated than that. Saari has made a career out of pushing the Borda count. There are useful applications for it, but I pretty firmly believe public elections are not

    It's a pity that Condorcet is ignored here, because he was da man. Condorcet's method kicks butt when compared to Borda and Approval (Approval is simpler to implement, though).

    There's a whole bunch of links to articles like this one in the Voting System category in Netscape Open Directory.

    Rob