Data Comparisons

Dynamics of online polls. (Weblog awards!)

67 Comments

The dynamics of online polls fascinate me. The limitations of inexpensive online voting technology, rules that permit repeat voting, and social dynamics lead to some fairly predictable behaviors.

Since blogs capture comments as things unfold, we can watch the process as it evolves. Here are select comments from Pharyngula interlaced with some from my blog.

  1. Voting in the Science blog category opened sometime after 3:pm central time on Jan 5.
  2. By 5:37 PM whatever time chris announces his plan to figure out if he can cheat:

    Posted by: chris | January 5, 2009 5:37 PM

    You can only vote once every 12 hours. That makes poll crashing not as fun or fast. I am using firefox (take that ie) and cleared my private data. It didn’t work.
    Next I will use crap cleaner and see if that works. If anyone has workaround please post.

  3. Soon, chris announces that he has figured out how to cheat.

    #35 Posted by: chris | January 5, 2009 6:28 PM

    Oh yeah I forgot to post. My bad
    Yes the free program cc cleaner cleaned whatever cookie or thing the site uses and i can vote again

    Oh another thing this is my 2nd post. Often lurking not posting.

    (BTW: I have no idea if Chris’s method works. It may be he hung up and his IP address changed or something else happened. Update: Sean Gleason from the Weblog awards reports they are on to this method. They can detect that it has been done, and if you do it, none of your votes will register.)

  4. People discuss whether or not one should cheat. There is a fair amount of argument by name calling.
  5. Meanwhile at my blog, I make a prediction.

    lucia (Comment#8126) January 5th, 2009 at 10:37 pm

    MikeC– Anthony is pulling ahead. I anticipate Boris is going to be disappointed. I suspect Anthony’s commanding lead over Steve will cause their mutual fans to vote for Anthony. The skeptic/lukewarmer vote won’t be split.

  6. Back at PZ Myers blog, PZ tells people not to cheat, does not erase Chris’s convenient instructions and reminds people cheating happened last year.

    Posted by: PZ Myers | January 5, 2009 11:39 PM

    Yeah, don’t cheat. The webhosts review the votes, and throw out any that look suspicious. Remember last year? There were some major revisions at the last minute as the dups were tossed out.

  7. At my own blog at 7:18 am, I post

    Given that people get very upset about these things at blogs and forums and then they start trying to drum up votes and voting continues, Myers could overtake Anthony in 24 hours.

    Plus, we can’t overlook the possibility that weblog awards might not have learned how to harden their voting script, and vote-bots could take over the race.

    Hey! I correctly predicted the someone would suggest what comes next!

  8. Posted by: thingsbreak | January 6, 2009 10:32 AM

    The anti-woo crowd needs to rally around one pro-science blog rather than split the vote, as the climate troofers seem to be swarming to Watts Up With That.

    (New to me word alert. I had to look up troofer. I googled for a definition of “woo” and found this use of woo which would so suggest pro-woo is has something to do with non-science based medicine. )

  9. Brain D reacts to thingsbreak’s suggestion:

    Posted by: Brian D | January 6, 2009 10:56 AM
    [. . .]
    While I wouldn’t classify Pharyngula as a science blog first (since most of the postings have to deal with atheism), there is still more science content here than there ever has been on ClimateAudit. Although it wouldn’t have been my first choice, ThingsBreak raises a good point: Pharyngula will receive my votes (one a day) as a rally for science

    So, to block other contenders, someone who does not consider Pharyngula a science blog has will vote for Pharyngula as Best Science Blog. This happens in real life politics; naturally it’s going to happen in online polls.

Prediction for today: Even if Chris’s method works, no one older than 14 is going to want to constantly clean cookies off their machine just to revote over and over. Those under 14 won’t care about this blog contest. Even with Myers evidently large following, too few people will read thingsbreaks suggestion to affect the vote count. Anthony’s lead will hold for at least 24 hours.

Prediction for tomorrow: If Chris has identified a real flaw in voting, anything goes!

Long term prediction: This blog contest will have zero impact on scientific understanding or policy decisions. It’s just good entertainment.

For those who do want to watch the races, I’m adding links to races involving blogs that discuss climate with some frequency.

Poll Status

  1. Best European Blog (Non UK) are open. Lubos of The Reference Frame has taken an early lead.
  2. Best Online Community Jennifer Marohasy lags.
  3. Best Science Blog Anthony Watts, PZ Myers is in second.

67 thoughts on “Dynamics of online polls. (Weblog awards!)”

  2. Lucia: Watts is miserable. I could write a book about what that guy doesn’t know. Seriously how can a good fluid mechaniker have any truck with such a lightweight?

  3. TCO–

    Anthony was nominated. I told people. He is ahead. I tell people.

    You spend a lot of time in comments. Why don’t you start your own blog, write your chapters and post them as a serial?

    BTW: I finished the afghan. I’m starting a new one. It’s also out of red heart. They each take 70 ounces. If you’d to send me Koigu, I’d be happy to use that instead of red heart. (It might cut into your Drambuie budget.)

  4. It would be one of those blogs with 0 comments below every entry… in the mean time, it’s bedtime in the Central time zone and Anthony is up by 600

  6. The cheating method proposed by Chris on Pharyngula will not work. It will succeed in allowing the client to send new votes, but those votes will not be registered. In fact, since any client caught attempting to cheat has its votes deleted, this method actually reduces the user’s number of votes from 9 (that’s one a day) to zero.

  8. Pingback: My picks for the weblog awards « Watts Up With That?

  9. Sean–
    Cool! I figured after last year, you guys would be sitting down,thinking of things people do and figuring out how to catch them!

  12. This is so facinating that I’m having Fruity Pebbles and donuts for dinner so I wouldn’t have to cook…
    so at dinner time Anthony is up by over 1,100

  16. … as the Rice Krispies pop, Anthony’s overnight vote continues to pay off… his lead nears 1500
    A big race now is the Conservative blog, Small Dead Animals is surging after heavywieght public endorsements like Anthony Watts, Kristen Byrnes and the Anchoress
    Lubos continues to ride high… Jennifer continues to struggle

  17. I’m just going to look at the % now. Anthony is in the lead with 36%; Myers is at 24%. These seem stable. So, unless voters get tired, Anthony is likely to win this. Of course… you never know!

  18. If nothing else, this always gives some folks in the climate science community a good incentive to rethink their poor communication, given how well the skeptic blogs do every year.

    That said, its really no surprise that Anthony is doing quite well in the votes, given that CA tied for first last year and WUWT has taken up the mantle as the more accessible destination for all things climate skeptic.

    Pharyngula can make fun reading at times, but unlike the climate debate picking on creationists is a tad too easy :-p

  19. Zeke–
    I wish the result of the polls did cause those in climate science community to rethink how the communicate. More imporantly, I wish they would decide to stop listening to all of Susan Hasol’s advice on how scientists can improve communication! (here.)

    Some people are soooooo obviously trying to follow it, and the results is to make the audience think the Hassolized communicator is trying to sell a used care with concealed body damage.

  20. Wow, Lucia! Thanks for the Eos link.

    Somehow, this reminds me of the occasions where a consumer advocacy group gets a hold of an article from a trade magazine. The insurance industry usually victimizes itself, where some wag will write an article in a trade mag advising insurors how to maximize profits by using clever sales tactics or by minimizing payouts. Or maybe a pharmaceutical trade mag where someone expounds eloquently on how to get Doctors to push pills onto patients.

    Here, we’re told climate scientist just need to talk to the peasants like they’re a bunch of idiots. Yeah, that’ll win ’em over!

    Look at this gem:

    “Error” means wrong or incorrect,
    so referring to error bars sends the wrong
    message.

    I wonder how she feels about the word “fingerprint”. 🙂

    After all, that might connote the criminal element.

    This is another doozy:

    Similarly,
    to the public, “uncertainty” generally means
    we do not know if something will happen,
    so uncertainty about future warming is
    taken to mean that it might not warm at all;
    it might even cool, for all we know. But that
    is not what scientists mean; they mean there
    is a range of possible warming, depending
    on the level of emissions and how sensitive
    the climate is to those emissions. So instead
    of “uncertainty,” try using “range.”

    Oh no-o-o-o. We can’t send out the message that the temperature might actually drop. What does that do to Gavin’s argument wrt to your trend tests? Isn’t his whole argument based on “uncertainty”?

    Really, all too typical. Talk to the common folk like they’re a bunch of idiots. After all, they only pay the taxes that fund the research.

  21. Johnm–

    In fact, the “range” in the context of that test includes realizations where temperature trend will be negative. Evidently, this could last quite some time now.

    So, yes, the uncertainty interval includes temperature include realizations where the annual averaged GMST will drop.

  22. Zeke, Lucia, you can make all the excuses you want for what’s happening here but you’ll never get over the hump… not as many people are believing AGW as some would like to think. That has nothing to do with communication, it has to do with who winns the debates and who shovels more snow and when, to use some examples.

  26. Mike C. Yes– 36.6% of the vote! BTW— Our 7 day forecast is snow, snow, snow, snow, snow, snow.

    TCO– No comments with the F word or discussing your desire to make babies. They are gone, gone gone. Not just Troll Controlled. I think i deleted the youtube too.

    Malabrigoyarn is beautiful. But I only make socky-slipper things out of El Cheapo. I started the blue afghan in Red Heart.

    I skimmed a bunch of comments, time for bed. My feet will be toasty warm in my ack-rylic sippers!

  27. I wonder if WUWT and CA can crack 50% of the votes between them? we are currently on 49.4%. I also found the most hilarious insult for AGW skeptic yet!
    Apparently we are a bunch of climate deniers! LOL, we don’t believe in climate 🙂

  28. Rob–
    The Weblog awards let me embedd a poll! Initially, they said only finalists can embed to avoid server overload. I agree that would be an issue, and suggested they created a version that is cached hourly. But anyway, about 24 hours after I suggested the caching, I got a poll! 🙂

    Maybe they now cache, maybe not. I’m not sure. Either way, we can watch the election returns– just li,e the presidential election.

  29. I’m voting for WUWT this year. CA won last year, is sometimes too technical, and sometimes gets into personal vendettas, whereas WUWT is always very clear, and always cheerful and polite regardless of provocation.
    Bob, another term I find very amusing is ‘climate legislation’, see the latest Revkin DotEarth article, as if we can legislate that temperatures aren’t allowed to go up more than 1 degree – it reminds me of the King Canute story.

  34. Jen seems to be competing against what looks like flicker photo group. I’ve never really visited those, so I didn’t know they had a blog aspect.

    When is this over? It looks like Anthony and PZ’s percentages of the vote are holding fairly steady. So, it’s looking like Anthony will win, as will Lubos.

    SteveM looks to be in 3rd with 13%, and RC has around 4%.

    Taken together, climate blogs have over 50% of the votes.

  38. MikeC–
    I think Pharyngula’s percentage has risen enough to show 25% instead of 24% for the first two digits. This doesn’t seem to be coming at the expense of Anthony– it’s coming from non-climate blog voters.

  43. 2800…
    and last year’s results are not going to tell you anything because of the cheating by the AGW side. I mean, come on, they snuck a bot into the awards system and forgot to program it to turn off at closing time.
    2006 might be a better year to look at, besides, RC didn’t do any better in 06 than they are this year. It might be time to start asking if RC isn’t really just a PR stunt with no popular following (kind of like saying the IPCC is made of 2500 or 4000 scientists). Just another big lie being pushed by a political PR company.

  44. MikeC,
    RC does have real readership. For some reason, their readers don’t seem to vote for them. Maybe they are busy; maybe they are just casual readers. Maybe they all decided to vote for Pharyngula.

    Who knows?

    I think these online polls are fun, but it’s hard figure out what either low turn out or high turn out might mean.

  45. Lucia, In 3 years RC didn’t even get nominated in one year, in the other two years they get a few percent of the vote, tops. If they have real readership then it would be the skeptics going to read what they have to say. PZ Myers group seems to have a steady vote and they are an anti-religeous niche. He scored fewer votes last year because he asked his voters to back BA in the last few days.

  46. Mike C,
    I don’t know why they didn’t nominate RC. Who knows how they did it? RC does seem to have quite a bit of readership. We do know that when they are nominated, the don’t draw all that many votes. They have less than 4%– whatever that means.

    For some reason, the way they opperate their blog doesn’t draw a lot of avid fans. I could speculate why– but who knows if my speculation would come close to correct?

  47. MikeC,

    The AGW types insist the bot was part of a “vast right wing conspiracy”. Is there objective information somewhere that would settle who was to blame for the bots?

  48. Raven, consider the logic… the vast right wing conspiracy installs a bot to defeat the skeptics so they can cry foul? Rather than just win the award? Pardon me while I cough up a bowl full of jelly flavored laughter.
    Licia… to create an impression that is not real.

  50. Mike C,

    Bots were used by both sides. The question is who started the bot war. The alarmists claim it was the “vast right wing conspiracy”. You claimed it was the alarmists. Does anyone really know who started it?

  51. Raven, you claim that bots were used on both sides, it didn’t look that way to me… I was watching that race as close as I’m watching this one. Last year Steve’s supporters were campaigning way harder, many blog endorsements including Junk Science and etc, hittinng up mailing lists, text messaging and etc. This year Anthony is doing so well that the same people decided to stand on the sidelines.

  52. Mike C,

    I was not watching the poll last year. I got the impression that they declared it a tie because both sides were using bots. If only one side was using bots then why didn’t they disqualify that side?

  53. I can comment on last year’s. CA was nicely ahead when voting closed, but Bad Astronomy got about 2500 votes after hours, so there was no doubt that bot voting had happened on their side. Having said that, there were about 4 times as many votes for each party on the last day as in the entire competition. It was exciting to watch, but, in my opinion, it would have been absurdly complicated for the organizers to try to sort it out. Both Bad Astronomy and CA had been running evenly up to the last day. I sent a late email to Phil Plait suggesting that a draw would be in the spirit of the competition. I also suggested that the votes be arbitrarily set at a round number so that it would be clear that these had been agreed to and was not just a coincidence. Plait agreed somewhat grudgingly – I don’t think that he really understood that he had got about a couple of thousand votes after the contest closed. I communicated this to the organizers who were absolutely delighted with this resolution and adopted the solution in about 1 minute.

  54. Steve–

    Yes. I remember that! I remember commenters reporting sudden batches of hundreds of votes appearing as if by magic on the last day.

    I can’t imaging WUWT not winning at this point.

  55. And the difference last year between CA and BA was that the CA votes were following a predictable pattern. When a big blog would endorse him, the numbers would start climbing until the story fell off the lead of the blog, not just jump like the other side. And the votes for BA climbed after voting closed, it was bizarre.

  56. This year, the vote patterns have been pretty much locked in stone since day 1, with ratios of remaining within a very narrow spread. So much so, that, In a way, each day has been a sample. If the last day this year is hugely anomalous, then I think that the blog organizers would have to assume that their system had been breached somehow.

  57. Raven, because theur policy was to throw out the bad votes, not disqualify the blog.
    Steve, If you’ve been watching Ace and SDA you would understand how the blogs themselves have been affecting the voting this year. It’s all been good.

  58. SteveM–
    I agree with you.

    Mike– Yes, Ace and SDA’s campaigning can affect voting, but there is a limit to how quickly people can spread the word. What happened last year was faster than what Ace and SDA are doing. Someone would pop into comments and say “Wow! 342 votes for CA in the last hour!”. Then someone else would say “Wow! 313 votes for blog BA in the last hour!”

    Pharyngula is well behind Anthony. But suppose he were in the lead. I know Myers upset lots of Catholics by blogging about desecrating hosts. How quickly could any group of Catholics organize together to vote against Myers?

    I know I could send an email to my Uncle, and he devoted to sending out email about this sort of thing to zillions. But would everyone read it? Vote? Forward it? Would those people vote? I have an email filter for my Uncle. The emails get stored, then in my own sweet time, I read them. So, the one he sends to me won’t get timely action. I’m sure there are people who just ignore the emails. So… How fast can anyone get out zillions of vote?

    Science *already* gets more way votes than Law. How many more people would bother to vote? (BTW The Volokh Conspiracy was in the lead last time I voted.)

    I doubt if we’d see 10,000 votes in a day as a result of some email lists.

  59. Lucia, your speculation will not change observed events.
    But while we are on the subject, PZ Myers cut Anthonys lead by 500 overnight.

  60. MikeC– You are correct that speculation doesn’t change observed events! Yes. Pharyngula got a lot of votes overnight. They are up to 27%.

  61. Christians are entitled to encourage folks to vote just like any interest group. But unless someone has seen a newsletter/Sermon urging folks to be behind Watts, I see no evidence they care for any of the candidates.

  62. Sean — Sure. I see no evidence this is happening.
    What I mean is that I’m not sure how effective campaign might be in this race. Hypothetically, they could work. But as a practical matter, could such a campaign work?

    I’m not sure. I also haven’t seen any evidence that anyone is trying to do this. I would imagine that few Christians have been staying up nights last week worrying about whether or not Myers might win a blog award!

  63. SDA is the only conservative blog that has been pushing Anthony. By the way, it looks like the surge of votes for Myers was created by his own blog posting. Now that Anthony has responded, the change in lead is slowing down.

  64. Well, now I don’t have to email my Uncle Bill, and tell him whip his elderly Jesuit educated high school and college classmates into a frenzy to vote against Myers! 🙂

  66. Grats Anthony!

    Even though we may disagree vehemently at times, I can’t help but respect the community you’ve built up over the past two years.

    Lucia,

    Look what we find over at Realclimate today: http://www.realclimate.org/index.php/archives/2009/01/communicating-the-science-of-climate-change/#comment-109123

    “We’d also like to alert our readers to an insightful article on this topic published last year in Eos by Susan Joy Hassol of Climate Communication (linked in our blogroll for those who want to learn more about the organization)”

    I think you called that one…

  67. Thanks Zeke!

    My husband gets EOS and brings it home. When I read the Susan Joy Hassol of Climate Communication article in EOS, I had an “Aha moment”.

Comments are closed.