« A splitting headache |
| Divided nation »
Via the Data Mining blog, I came across ChoiceRanker, which is some kind of straw polling site that uses the following visualization of the data.
This particular chart is related to the question of who should be Obama's VP. I can't say I understand what's going on here. If you can figure it out, let us know.
Posted on Jul 05, 2008 at 11:44 PM | Permalink
TrackBack URL for this entry:http://www.typepad.com/services/trackback/6a00d8341e992c53ef00e55389c2e48833
Listed below are links to weblogs that reference It's raining colors here too:
Round 1 has votes for all candidates. Then at each round they eliminate the least popular and the lower row for each round shows how their votes are divided.
Jul 05, 2008 at 11:56 PM
The ballot lets you rank candidates in order, 1-10. Say you ranked Hillary Clinton #1, Jim Webb #2, John Edwards #3, and some other candidates #4-10. When Hillary was eliminated in round 1, Jim Webb got your vote starting in round 2. When Jim Webb was eliminated in round 8, John Edwards got your vote in round 9. This was enough to give him a majority of the remaining votes, so he won.
Proponents of IRV say it's a better system than plurality voting because it's better for third parties. Say it's 2000 and you support Ralph Nader. With IRV you can vote for Nader and not be afraid you'll throw the election to Bush. That is, you don't have to vote for Gore for fear that not enough other people will vote for Nader.
Sure this chart has a lot going on, but it would convey a lot of information to someone who understands the system. I don't, however, think it's the best advertisement for IRV.
Jul 06, 2008 at 12:55 AM
There is a video that explains it, but nevertheless, if you know in advance what Instant Runoff Voting is you can guess it.
As Ken has said, the bottom graph shows the distribution of the first preference votes. If nobody has 50 percent of the vote, the candidate with the least votes is eliminated, and the second preferences of his or her voters are counted. In the round 2 graph, you have, then, in the upper part, the votes for the remaining candidates after adding 2nd preferences, and in the lower part, the composition, in terms of votes of the previous round, of those votes.
You do the same thing again as long as no one has got 50 percent. Each round, you eliminate the candidate with the least votes (which include 1st, 2nd, 3rd preferences...), and add its voters to the next preference (2nd, 3rd... whatever).
The black bar in the right accumulates the non-votes of voter who voted only for a few candidates (you are not obliged to rank all of them) and all their choices are eliminated. In this case it is big, because of the number of candidates. In real life it would probably be much smaller.
You stop the process when one candidate has more than 50 percent of all votes (round 9), or when you have only two candidates left (round 8), which produces the same result. Again, in a real election these two things happen at the same time, but here, because of many non-votes, they happen at different rounds.
In this case
Jul 06, 2008 at 06:33 AM
Except for the glaring colors, I had no problem with this chart. Of course, I'm familiar with IRV, since we used it for elections in college.
Jon Peltier |
Jul 06, 2008 at 09:14 PM
This note is from Craig Simon. I created the site you're talking about.
FYI, I call the main graph you're talking about an "Irving Chart." It's good to know that folks who understand IRV were able to figure it out.
For newcomers, there is a 2 minute video that explains how things go. I'm confident that most people can understand it, because my mother (who is not particularly computer savvy) got the main idea in less than a minute. For more sophisticated observers, there are tooltip pop ups that convey lots more information (Labeling got way too busy).
The bar charts on the left side -- I call them "depth" charts -- show a modified Borda count. These work much like college football rankings, but are more heavily weighted to a voter's 1st place vote. The depth results are used to break any ties during the IRV rounds.
Also, no one actually "wins" till they get 50% of the total votes cast. That's why it may appear as if there's an extra round after it gets down to just two candidates. That final distribution provides a better sense of whether the winning candidate is acceptable to everyone.
Also, I'm not seeking to promote IRV; it's a means to end. The goal is to create a system that can help diverse communities reach a consensus.
Craig Simon |
Jul 06, 2008 at 10:57 PM
The video on their website actually does a good job of explaining it. It this chart didn't have any labels, I would have thought it was an application for gene sequencing.
Jul 11, 2008 at 02:04 PM
This is only a preview. Your comment has not yet been posted.
The letters and numbers you entered did not match the image. Please try again.
As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.
Having trouble reading this image? View an alternate.
(You can use HTML tags like <b> <i> and <ul> to style your text.)
(Name is required. Email address will not be displayed with the comment.)
Name is required to post a comment
Please enter a valid email address