Oh, that's gotta hurt

The results are in, and even though Joe Romm suggested:

Please click here and freep this poll until the magazine has the decency to take it down.”

He apparently had little effect. I do agree with Joe though, the poll was poorly designed. For example look at this result:

Only one problem, the math for percentages doesn’t add up:

That’s because the poll allowed for more than one answer to some questions. When you do that, percentages then don’t reference properly to 100%.

As I and many others said, it was poorly designed and poorly presented. From my vantage point it looked like very little thought went into it.

That said, the results, while interesting, should be taken with a grain of salt.


106 thoughts on “Oh, that's gotta hurt

  1. Gonna be tough to hide the decline of the percentages who believe in AGW….err…climate chan…….ahhhhhh whaterever!!!

  2. Even if you could vote for more than 1 answer, the number for “skipped question” should have just kept increasing. The bigger number of skipped questions I saw was 40 (early in the poll), then it was always kept small, in the single digit.
    Climate scam = polling scam

  3. Wrong but probably not deliberately misleading at the math level.
    Its the interpretation of statistics that leads the the greater depatures from the truth

  4. Ahem….. it is widely known that 67.49% of all statistics are made ‘up on the spot’. Reference Mann/Jones et.al. statistical methods for examples.

  5. “Please click here and freep this poll until the magazine has the decency to take it down.”
    “Freep” is a term used by the conservative forum http://www.freerepublic.com to swarm a poll. Somehow, I find it hard to believe that Joe Romm is a FReeper.

  6. I found the results amusing.
    (I especially find it amusing that journalism ratings seem to indicate that AGW is so last year.)

  7. The real story is that someone at Scientific American doesn’t think that solar variation is a natural process.
    Given that they are clearly thought leaders on every front and are in on the truth, let me be the first to welcome our new Sun manipulating alien overlords.

  8. I would have thought that scientific theories, observation and (dare I say it) commonsense, would mean that the first three options would all have to be checked (ticked).
    The only question is – how much of climate change to allocate to each factor? Observation of the effects of El Ninos/La Ninas (alone) would indicate that the relative allocations themselves would vary from time to time.

  9. You can count on the AGW proponents to forget simple math… that’s why the climate hasn’t followed their computer predictions… LOL

  10. I actually liked it.
    And if they continue to do things like that, and publish real (sceptic) science, I would consider reading them as well.

  11. Only one problem, the math for percentages doesn’t add up

    But this is peer-reviewed climate science, Anthony! It uses different laws of mathematics to the rest of science.

  12. It’s not a true scientific study and I’m sure there’s bias one way or another in who answers the questions. (The sampling in NOT unbiased.)
    However, it’s interesting to see that the majority don’t buy into the greenhouse gas theory.

  13. Re percentages. They won’t add up, not I think because they’ve got it wrong, but that a good proportion of those who answered yes to natural processes would also answer yes to solar variation as a subset of natural processes.

  14. The option I would want is “No reliable information available to draw a conclusion.” Given that the majority of the climate science community has defended the publishing of graphs that hide the errors of proxies by replacing proxy temps with thermometer temps, no evidence brought forth by that community can be trusted.
    I wonder if anyone is going to do a confirmation of the study by Edward Long, where he compared a rural and urban station in each of the fifty states and found a divergence after about 1960. It doesn’t seem like it would take to long to go over a hundred stations and verify that he didn’t cherry pick. His study seems like an extremely strong indication of error from the urban heat island or similar effect. There seems to be very little discussion of his study on the net. It doesn’t seem to have been debunked either. The study is:
    by Edward R. Long, Ph.D.
    And Anthony, the release of your surfacestations study is taking so long that it is starting to look like you are hiding something. At least give us an update.
    REPLY: The paper is written, submitted, and in peer review. We’ll give a full release of data when published. It is out of may hands now, waiting on the journal – Anthony

  15. I think the SciAm editors were just having a little fun and building some buzz. Online polls are inherently unscientific anyway.
    From the Romm viewpoint the current numbers are “worse than we thought” as of his posting. One thread commentator stated: “What happened here is that the deniers posted this poll on a website like WTFUWT and people went in droves and distorted the balance.” However I found the poll and answered it before it was posted here and even before the WUWT effect the results were pretty bad from an alarmist perspective; they just got worse after.

  16. We humans breath out CO2. So do all other animals. Therefore, we and all other animals must die, to “save the planet”.
    We are living in an era not matched in history for mass hysteria.

  17. “The results, while interesting, should be taken with a grain of salt.”
    That statement could apply to myriad aspects of climate ‘research.’

  18. Those grains of salt will come in handy. As seasoning, as the (C)AGW alarmists are slowly roasted over the hot coals…
    Hey, they’ve been feeding off the public for decades. Turnabout is fair play. ☺

  19. The first sentence has a fact in error. Judith Curry is at Georgia Tech instead of the University of Georgia. These are who we turn to for scientific facts?

  20. Anthony your criticism of the percentages is misplaced. A few of the questions allowed for multiple answers.
    REPLY: May I suggest that you fully read the text? That’s exactly what I said. Then read this definition of percentage

    A fraction or ratio with 100 understood as the denominator; for example, 0.98 equals a percentage of 98.

    That’s the “cent” in percentage. 100. Lesson: don’t use percentages to show results if they don’t work with adding up to 100.

  21. Yeah – isn’t that the fate fashionable ideas…I guess CAGW is going the way of
    love and peace;
    pet rocks;
    long/short hair/skirts;
    immanent ice ages;
    hope and change;
    fill in the blank with any other silly faddish thought fashion: ________________

  22. The idea of multiple answers doesn’t bother me. Sometimes polls make more sense that way. In the example question that you highlighted, “Natural processes” would include “Solar variation.”
    Nevertheless, Scientific American has a long way to go before they regain my confidence.

  23. Anthony – only a GRAIN of salt?
    I took the survey and could not complete it untli I had answered ALL questions – so how did the 2 ‘skipped question’ respondees manage that?
    There were a couple of questions that I did not like any of the options, eg No5 – ‘What should we do about climate change – as we do affect out local climates through construction, deforestation, etc there are things we can do but the choices to the question are too broad in scope.
    It is interesting to see so many opting for ‘Keeping science out of the political process’ as politicians have to be guided by scientists for many of their decisions. The option should have been ‘Keeping POLITICS out of the SCIENTIFIC PROCESS’.

  24. LarryOldimer says:
    November 5, 2010 at 1:31 pm
    We humans breath out CO2. So do all other animals. Therefore, we and all other animals must die, to “save the planet”.
    And cows fart Methane, adding insult to injury. I think, therefore, they must die. I’ll be in linwe for the steaks.

  25. Interesting, indeed. And I’m sure many skeptics refused to take the poll due to the lack of available answers accurately reflecting said skeptics’ views.

  26. Perhaps this is the latest in Principal Component Analysis? Only the second component is completely included in the third. (That is, unless solar variation is not a natural process.) Anyone choosing the second option should also choose the third. And if you deduct the number choosing option 2 from that of option 3, you do begin to approach 100%. (With characteristic SA accuracy…)

  27. Actually, Anthony, I thought the provision for multiple responses was a strength of this poll. I voted for three of them, because I do believe that both natural process and human agency are involved and I was quire prepared to add solar variability (yes, that too is a natural process and should have been given as an example of natural causes). So the poll did not try to force a false dichotomisation of the causes, which is surely a virtue.
    REPLY: I have no trouble with multiple choice, but showing the results in percentages is the wrong way to do it. [per”cent” = fraction of 100] All they needed was the bar graph and totals, or they could have broken up the questions so that the answering process was better defined. – Anthony

  28. Exxon 1,614
    Round shiny thing 1,745
    James Hansen 3,982
    What’s climate? 325
    = 7,666
    They say 5,258 voted. If we take away solar (because it’s natural) we get 5,921 which should be robust enough to pass Nature’s vigorous peer review process.

  29. Hmm,
    So, as a percentage of answers given, we get
    So when attributing climate change, natural variation gets over half. How does such a smart readership put up with that amount of editorial idiocy?

  30. These numbers prove there is a consensus of scientists (readers of Scientific American)
    that climate change is caused by natural processes.

  31. OMG! Just looked at the whole poll. Yeah it hurts! Hurts from laughing! Doc, this has got to be – – a joke? a ruse? an elitist’s put-on? a con to see the reaction?
    but only via serious venues like The New York Times ROTFL!
    an effective group of government… OTF – still laughing!
    the phrase on which the fate of human civilization hangs My side hurts! Stop!
    If it’s not a joke, it is the absolute worse poll I have ever read.

  32. My take on it… 79% of the votes cast agree that “greenhouse gases from human activity” have no influence on the climate.

  33. and in # 7, keeping science out of the political process
    Shouldn’t that be; keeping the political process out of science??
    at least as another option?

  34. Alert to Josh!!
    SA is going into comedy. Based on their hilarious survey they are serious competition for you.

  35. Sorry Anthony, I was out of line to even suggest that you might be hiding something. I just want to see your study so badly that I’m getting frustrated. But I should remember that the reality is that things often go much slower than it seems like they should.
    [REPLY – Patience . . . ~ Evan]

  36. Anthony, I doubt Joe Romm wanted it taken down because it was “poorly designed”.
    No, I’m just glad that “solar variation” at 33.2% and “natural processes” at 74.7% add up to 107.9%!
    If the “warmers” can play with numbers, why can’t Scientific American?

  37. So much for Scientific American. A clear demonstration of their scientific prowess.
    What will we see next? A big article on how the earth is warming and we are all going to die?
    Oops, they already did that.

  38. 8. How much would you be willing to pay to forestall the risk of catastrophic climate change?
    nothing 77.5% 4,215

    muahahaha! And that’s the truth of it. 😉

  39. I imagine the numbers are homoginized. What you need to see is the raw data and the code used to calculate the results.

  40. I wouldn’t worry too much about the stats. I’d just see it as saying:
    1. No-one thinks there is no climate change. Good, because that’s true.
    2. There’s far more folks thinking climate change is natural or solar in nature than AGW. Good, because that’s the most likely long-term truth from the scientific data currently out there.
    There’s no need for this nitpicking.
    The results are the way you want them.
    So why nit pick?

  41. Isn’t ‘solar variation’ a member of the set ‘natural influence’?
    Scientific American has fallen a long, long way.

  42. I thought it was a fun, entertaining set of questions.
    The thing I thought was the most entertaining was trying to figure out the mindset of the crew that thought up both the questions and the possible responses.
    That mindset was clearly bizzare.

  43. How can you be sure that loads of Joe Romm’s readers did not accept his advice?
    Perhaps the majority of his readers only go there for a laugh!

  44. Mindbuilder says:
    November 5, 2010 at 2:44 pm

    Sorry Anthony, I was out of line to even suggest that you might be hiding something. I just want to see your study so badly that I’m getting frustrated. But I should remember that the reality is that things often go much slower than it seems like they should.

    Hi Mindbuilder,
    I don’t know if you’re worked with peer-reviewed journals in the past, but often times things move at a pretty slow pace. You submit the manuscript, things get processed in a couple days or weeks, then they get sent out for review where they often sit on a prof’s computer or desk for months before getting looked at. It’s a slow process. I work in a “fast” field, so I’ve had papers published online in ~2 months, but I’ve also sent them to other journals where even the first reviews didn’t come back for several months.
    I can’t imagine how much longer it’ll take for a manuscript like the surface stations one to get reviewed. All it takes is one reviewer (or editor) who doesn’t like it to drag out the process to well over a year, even if the majority of the paper is fine. And the more feathers you ruffle, the longer it takes. Unless you’re lucky, the only way to get a paper through faster is to publish something that doesn’t upset anyone, and that’s not going to happen with the surface stations work.

  45. actually, aside from the labels on the graph, I’d say that multiple answers is a good polling structure for a question like this. Perhaps “Respondents answering positively” and “percent of respondents answering positively” would have caused less confusion.
    Gotta be encouraged by the results, all the same.

  46. You’d be better off keeping that salt for de-icing the roads in the coming northern winter – I suspect you are going to need lots of it.

  47. I had to look up freep:
    “To subject to a mass internet or email assault aimed at pushing a particular point of view”
    Disturbing to learn that google is now also a dictionary. Sadly not surprising that Romm was encouraging such childish tomfoolery.
    Please add the definition to the article.

  48. Here’s the latest at 0035 hrs GMT+1:
    “What is the IPCC (paraphrased)?
    An effective group of government representatives, scientists and other experts. 17.0% 948
    A corrupt organization, prone to groupthink, with a political agenda. 82.3% 4,580
    Something to do with Internet protocols. 0.7% 37”
    Well, all I can say is that we are not all barmy.

  49. Anthony:
    I see no problem with the summary of the results as %s. As I am sure you recognize, for questions with multiple responses, the % simply refer to the % of those responding not of all responses. The summary of the results should also separately display the results for those who identified one, two, three of the apparently four options. It would be interesting to know the % of people who attributed the perceived warming to solely greenhouse gases and those who perceived no role for greenhouse gases.
    REPLY: Well perhaps it is just the TV reporter in me, but we had a cardinal rule in the newsroom that ANY graphics that went to air or to web using a percentage had to be checked so that there was no math error or entry error and it all added up to 100%. This was particularly true of election season where we might have six candidates for one office and if the % figures didn’t all add up to 100, we’d get calls from all over the place.
    I remember well a math teach from a local school calling up and berating one of the neophyte reporters once for just such a gaffe. The poor girl was in tears.
    So from my perspective in presenting stats to the public, any graphic where the % of items doesn’t add up to 100% is just asking for trouble. I figure if you design a poll that allows such an presentation “error” to be automatically generated, you need a better poll design. This is the root of the issue here. – Anthony

  50. It is always difficult and dangerous to read too much into a poll like this. Many who are apathetic to climate change would not bother to vote. If one is apathetic, this is generally because you have doubts and are not a card carrying believer.
    Obviously solar variations is part of natural variation. The difference being that many drivers behind natural variation are not known still less understood (indeed how can you understand something that you do not know about). Solar variation is simply one of the known natural drivers in the all embracing term ‘natural variation’.
    Simplistically, the poll suggests a split of about 30/70 believe that climate change is anthropogenic, or put another way, approx 70% of responders do not consider climate change to be manmade. If so, that shows support in manmade climate change to be weakening when compared to other polls over the last 18 months or so. Good news indeed.
    Support for the theory will further decline should we have another cold winter in the Northern Hemisphere, and also in circumstances where people are struggling to make ends meet in the light of the western world recesssion and finding that their pain is increasing due to their energy bills going up by 10% because of green subsidies and holiday/air travel tax increasing. For a family, the rises in air travel tax in the UK will add between £70 to £300 on the price of a holiday. When disposal income is reduced, inevitably one questions these matters more.
    Further, slowly (perhaps I should say very slowly) the tone of reporting of events is changing. Today, the BBC ran a lengthy weather report (I did not see all of it) commenting on world wide weather events. It talked about record highs in Japan and wet weather in Mexico and said that this was all part of El Nino La Nina cycles!! There was no mention of manmade Global warming. They reported that this year was a very active hurricane season in the Atlantic with 17 named storms (earliest hurricane and first time 2 cat 4 storms had developed at the same time). Again this was attributed to El Nino (not AGW) and further they even said that this was not a record year, the record year being 2005 with 28 named storms. As I say I did not see the whole bulletin so may be there were AGW scares but what I saw was more objective.
    I think that many in the AGW camp see the real possibility of many years of cooler weather ahead, and realise that this will cause them real difficulties unless they tone down their message. I fully expect them to say that natural variation can for a limited period mask the effects of manmade factors such as increasing CO2, but despite this the underlying trend is still that temperatures are going up such that when the natural variations are less influential, AGW will come back and bite us.

  51. that poll must be wrong!
    Reuters: Despite Electoral Outcomes, Poll Shows Voters Want Clean Economy
    By Elizabeth McGowan at SolveClimate
    Those findings come from a survey of 1,000 voters who actually cast ballots in 83 battleground House districts nationwide. Washington, D.C.-based Greenberg Quinlan Rosner Research conducted the poll Nov. 1 and 2.
    When voters who chose the Republican candidate were asked to name their biggest concern about the Democrat, only 1 percent cited an answer related to energy or cap and trade…
    “There was no mandate on turning back the clock on environmental protection,” said Heather Taylor-Miesle, director of the Natural Resources Defense Council Action Fund. “Polls galore show continued and strong public support for making continued progress to protect our health and boost our economy.”…
    As well, 55 percent of those polled supported a comprehensive energy bill that charges energy companies for carbon emissions but also would limit pollution, invest in domestic energy sources and encourage companies to develop clean energy…
    By a 22 percent margin, battleground voters supported the idea of the Environmental Protection Agency tackling global warming by regulating carbon emissions from power plants, vehicles, factories and other sources. The poll showed 58 percent supported the EPA taking such initiative and 36 opposed the idea….
    “As sure as the sun rises in the East, America is going to continue moving forward on the clean energy economy and strong environmental protection,” said Anna Aurilio, director of Environment America’s Washington office, about the poll’s results. “The next Congress will have to decide if it is going to be responsive to science, innovation and public support or if it will simply focus on payback to Big Oil and the polluter lobby that funded so many of its campaigns.”
    look – even small business wants action!
    June 2010: New Poll Finds Strong Support for Clean Energy Policies Among Small Business Owners
    A new bi-partisan poll conducted by Greenberg Quinlan Rosner Research and American Viewpoint found strong support among small business owners for clean energy and climate legislation…
    •61 percent of small business owners agree that moving the country to clean energy is a way to restart the economy and help small businesses create jobs…
    even the Japanese want action!
    2009: Reuters: Japanese voters want tough climate goals – survey
    WWF and other Japanese and international organisations — which all favour tough climate goals — commissioned the poll of 976 people by U.S.-based Greenberg Quinlan Rosner from May 16 to 25…
    surely Sci-Am should get Greenberg Quinlan Rosner to design their next poll.

  52. Re Mikes comments on Rachael Maddow:
    I have to agree. We should put our complete trust in an admitted leftist, sexually confused anchorperson on CNN!

  53. Does anyone here try to get a comment published over at Climate Progress? I see almost ZERO contrarian views in their comments section.

  54. I didn’t renew my subscription about a year ago.
    Still a long way to go though before it reaches the appalling standards of New Scientist.

  55. Since Nature caused mankind to come about, isn’t “anthropogenic” global warming also coming from natural processes?
    You might be familiar with the legal reasoning where the maker of a defective device gets the blame for harm the device does rather than the device itself. Yup, old Joe Romm is going to keep at it until Nature regrets not issuing a voluntary recall. He’ll keep going until he gets one too! ☺

  56. The math isn’t so bad since people were free to give more than one answer.
    The % is based on the people answering the question, and is not based on the number of answer to the question.

  57. I have not read all comments.
    Lemonick must enjoy special privileges at SciAm; self-promotion of his article, by engaging a Monkey to offer a ´survey´ even more drivel-ous than his original effort. I agree, Anthony, ¨From my vantage point it looked like very little thought went into it.¨

  58. PhilinCalifornia says:
    November 5, 2010 at 5:30 pm
    “Still a long way to go though before it reaches the appalling standards of New Scientist.”
    Very true. Difficult to top these days. NS is toast.

  59. I agree that the polling is up the creek.
    The first problem is that the options aren’t exclusive. Anybody can choose all or just some. Even when one allows for those selective solar factors recognizing that they should also tick natural, the sums don’t add up at all (just ignore the count of solar) because some respondents probably thought that greenhouse gases also have some effect.
    Where’s no way of telling from here how many chose three out of four.
    One thing that the poll demonstrates is that SciAm doesn’t know how to construct a poll that produces meaningful results.

  60. Scientific American has really suffered since the loss of Martin Gardner….
    His mathmatical games were wonderful.
    But sadly, Scientific American has been forced to outsource their mathmatical problems to 2 individuals who, ironically enough, are brilliant in their venue, but their math skills are, shall we say, entertaining.
    Matt Stone and Trey Parker, who discovered ManBearPig, used the same Algorythm for percentages in one of their famous documentaries.
    ManBearPig is 50% man, 50% bear, and 50% pig. This new application of percentages is quite useful, though, and we can expect to see it applied by political parties soon.

  61. Anthony…I’d quibble a bit with your heartburn over multiple response questions. I’ve used them in some settings. Yes, you are correct that the poll was sloppy…for example, I looked at the original survey form and the question did not even state “choose as many as you think are relevant” (or something)….the only indication was a check box instead of a radio button (which is the forced choice format)
    The summary was very poor. But this is not an election. Yes, it would be horrible to go on the air with something that should add up to 100 percent but doesn’t. My favorite is the cume chart that starts sloping down somewhere in the middle.
    If properly explained….like “here’s the percentage of people who find the following factor important in climate change” there is information to be had.
    Still better would be for the respondent to assign weights….but that would be hard for survey monkey

  62. I reject completely the notion that the sun is part of natural variation. The sun is a nuclear device and as such is under strict regulation by the United Nations, the International Atomic Energy Agency to be specific. A considerable number of communities in fact have declared themselves nuclear free zones, though admittedly this was mostly about banning nuclear missiles (hence the signs erected at airports to prevent them from landing). Given the proven, robust, and obvious effectiveness of the United Nations in general (consider the fine work of, for example, of the IPCC, programs like Food For Oil, and the leadership demonstrated in the battle for women’s rights being chaired by Iran) the notion that there is any variability in the sun not carefully monitored and controlled by the UN is just nonsense, and anyone who says otherwise doesn’t understand science. All you have to do to prove it to yourself is look at all the IPCC models that don’t even include solar variation as a variable. If its not in the models, it obviously isn’t a factor. Sheesh, just read the reports! The proof is right there!

  63. davidmhoffer says:
    November 5, 2010 at 10:37 pm

  64. I don’t believe they’ve got the poll questions wrong. The one you point to Anthony asks what the cause of climate change is. I don’t believe anyone on these threads would deny that humans are having some effect on the climate, or that the sun is also having an effect. What this question tells you is that most people admit to there being some climate change from the humans and the sun, but the majority think that the major drivers for climate change are natural events.

  65. Well, one question if I recall correctly asked how much effort/money we needed to devote to stopping climate change.
    The last option was ‘Whatever it takes.’
    It was the only possible answer – so shrouded in ambiguity since ‘Whatever it takes’ could range from ‘absolutely nothing’ to ten years worth of sackcloth and ashes and abstinence from all that feels and tastes good.

  66. If you add the ‘human activity’ responses to the ‘natural causes’ responses there is an excess of about 328 which is close to the number of ‘no climate change’ responses….

  67. In the poll question I see:

    3. What is causing climate change?
    solar variation
    greenhouse gases from human activity
    There is no climate change
    natural processes

    Then further down the survey page it asks:

    5. What should we do about climate change?

    This is not a very scientific survey.

  68. “Isn’t solar variation a natural process?
    Last time I checked, yes. Unless the IPCC has declared otherwise…

  69. That graph at the beginning looks exactly like the the graph for heat absorption of gaseous CO2. Now there’s a coincidence

  70. Watts: “very little thought went into it”? I think they should regard this as Oreskes , or Doran/Zimmerman quality, and therefor publish it in Nature or in Science. 😉
    No! Just kidding. My humor is (just like AGW) bottomless…

  71. I’m puzzled why more folk here don’t appreciate the virtues of this poll. Sure, as science it’s a nonstarter. But it does reflect real swathes of public opinion, as one reads them in so many columns of readers’ comments where it’s plain the readers know far more, and are fairer and wiser, than the author of the article did or was. Moreover I think it was done with tongue in cheek. And surely it signals a real move in the skeptics direction from the NS op-eds and Mannian complain-o-logues we’ve seen even recently. Look at what SciAm readers now think:
    1. Should climate scientists discuss scientific uncertainty in mainstream forums? (resoundingly )Yes, it would help engage the citizenry.
    2. Judith Curry is: (resoundingly) a peacemaker.
    3. What is causing climate change? (clearly) natural processes.
    4. The IPCC, or Intergovernmental Panel on Climate Change is: (resoundingly) a corrupt organization, prone to groupthink, with a political agenda.
    5. What should we do about climate change? (clearly) Nothing, we are powerless to stop it.
    6. What is “climate sensitivity”? (good majority) an unknown variable that climate scientists still do not understand.
    7. Which policy options do you support? (good majority) keeping science out of the political process.
    I’ve been waiting years to see this commonsense returning.

  72. That’s the “cent” in percentage. 100. Lesson: don’t use percentages to show results if they don’t work with adding up to 100.

    Showing the responses in percentages for that question only shows what fraction (times 100) of the respondents thought that e.g. greenhouse gasses from human activity were a contributing factor. So they do add up to 100%… for each possible response the number who didn’t think that was a contributing factor was 100% – the listed percentage.
    It is erroneous that the sum of all of the columns should add to 100% before one can use percentages.
    A legitimate criticism might be that the question should have read:
    What is causing climate change (select all that apply)?
    Clearly there were people who were confused that only one choice was possible.

  73. Carrick says: “It is erroneous that the sum of all of the columns should add to 100% before one can use percentages.”
    Your point may be technically correct, but it is extremely misleading if they don’t. First impressions matter – people will look at a column containing a percentage and assume all of the items add to 100%.
    When presenting comparisons, results should account for the entire sample. Even if one of the items is something like “no opinion”.
    Bernie and sylvian say: “the % simply refer to the % of those responding”
    Then there should be two columns containing % agree and % disagree. The bars should be presented in similar fashion. The two will then add up to 100% and the presentation will start to make sense.
    I agree with Anthony 110% on this one!

  74. Lucy Skywalker says:
    November 6, 2010 at 8:12 am
    I’m puzzled why more folk here don’t appreciate the virtues of this poll.
    Agreed, Lucy S.
    I think it was a kite flown to see who would show up. Now they know there is a demographic that carries no truck wth their dyed-in-the-paper alarmism that has ruled for so long. The point of keeping it deliberately non-scientific was so the Romms of this world could rightously damn it and from which they could recover easily if it didn’t fly. I have no idea if 5k hits is a good response. They obviously keep track fo the number of re-ups for subscriptions and perhaps they are trying to staunch a bleed with a few alternative climate-medicine pieces. The New Scientist is bound to do the same though will probably take longer to turn things onto a more even keel being in hidebound Great Britain.
    Incidentally I visited the Romm site for the first time as a result of this (giving him a WUWT bump) and it was a bizarre experience. I think fulmination is the right word. It was worse than listening to Jerry Falwell. I literally think they do not hear the content of their own sentences. It is also clear, like other warmist sites, they revolve around the axis of WUWT and fear its frequenters and comments mightily.

  75. Just so everyone knows.
    A percentage can be based on the number of respondents OR the number of responses; i.e., N = responses or respondents.
    When multiple selections are allowed on a poll question, the percentages reported should always be with reference to the number of responses (not respondents) and it should be made clear which it is on each question. What this means is that the percentages should add up to 100% for each question no matter how the question is asked (single response or multiple responses per question).
    Now, if the individual percentages don’t add up to 100% for each question, it is an ERROR–usually by the programmer who implemented the poll on the Internet. The person who calculated the statistical percentages obviously used the “number of respondents” as the denominator on each question rather than the number of responses as he/she should have done. It was a common mistake caused by someone who, in a hurry, divided everything by the number of respondents. What is amazing to me is that no one at Scientific American caught the error.

  76. Stupid little people. Don’t they know how to respond to poorly designed surveys in the way they are expected to?

  77. Ideally, you ought to give the percentage giving each response and combination of responses separately.
    A: 40%
    B: 45%
    A and B: 10%
    Neither: 5%
    Instead of:
    A: 50%
    B: 55%
    Although I can’t believe anyone is really taking this seriously.

  78. Didn’t like question 7. I didn’t support ANY of the “policies” but they wouldn’t let me submit the poll without selecting one of the selections. So I selected “keeping science out of the political process.” That of course is the opposite of what I want. We need to get science in and Gore out. But I certainly didn’t support any of the others so that was the least worst selection.

  79. This reminds me of a humourous sketch by the Chasers,

    (well the bit 50 seconds in), and I think their statistics bureaus are very similar. Still, we should recognise this poll for what it is, an overwhelming proof that the consensus of the public is on our side no matter how they skew the presentation or results.

  80. In England, football (soccer) players are routinely encouraged by their managers to give 110%; so we are quite used to having more percentages than a paltry 100.

  81. Ditto what Lucy said.
    But what’s the meta-message here? Is SciAm engaged in exploratory self-correction? Is this a prelude to a CAGW flip-flop? If you were publisher, and saw your revenues decreasing and profits disappearing and sensed that your readership was deserting you, what would you do? We know they have eschewed real science for quite some time. This effort was political in every sense, not a return to science. Because that’s a touchy subject: if SciAm admits returning to science, then where have they been?

  82. From my vantage point it looked like very little thought went into it.

    It is an unScientific American poll. Why would you expect any thought to go into it. Si-Am has proven itself to be a garbage magazine with no credibility.

  83. Aynsley Kellow — and Anthony’s reply, concerning multiple answers to poll questions:
    In my view the reporting of outcomes to such a question is deceptive in any case unless answers are reported in grouped form. Nominally, since there are 4 binary choices there are 2^4 = 16 potential answers, though one would think any combining the fourth option with any of the first three are nonsense. Still I would be interested in the number selecting the last two. Persons who did this would be quite distinct from those selecting the first three, although both groups would contribute to option 3 in the statistics they reported here.

Comments are closed.