1. Welcome! Please take a few seconds to create your free account to post threads, make some friends, remove a few ads while surfing and much more. ClutchFans has been bringing fans together to talk Houston Sports since 1996. Join us!

That poll about Arab opinions on the US? Oops... Enron math...

Discussion in 'BBS Hangout' started by haven, Mar 24, 2002.

  1. treeman

    treeman Member

    Joined:
    Nov 27, 1999
    Messages:
    7,146
    Likes Received:
    261
    You people are amazing. It is so crystal clear that you're willing to grab onto any tidbit that will allow you to believe what you want to believe, and deny anything that might hint to the contrary.

    The poll was still valid within its framework and considering its parameters, and yet because it was reported wrongly in some presses you all just assume that it was flawed.

    The results of the poll are virtually unarguable. The way it was reported/represented (not by Gallup, FYI) is arguable. But the results stand, and they still support the general idea of "we are widely disliked throughout the Islamic world".

    Here were the methods used:

    http://www.gallup.com/help/FAQs/answer.asp?ID=70

    And a general link to the survey and its contents:

    http://www.gallup.com/poll/summits/islam.asp

    The main criticism here appears to be that Gallup didn't take comparatively proportionate samples among nations. Well, gee, I wonder why they didn't do that? I mean, Indonesia's got 200 million people, so a sample that was equally representative in comparison to Kuwait would only need to be about 6 million large... :rolleyes:

    Some of you don't appear to understand sampling techniques. You do not need to sample an entire population, or even 10% of it. Or for that matter, even necessarily 1% of it. You simply need a large enough randomly generated sample to approximate the population parameters - the number usually ranges from about 800 to 1400 sample size - in order to get an accurate approximation. It all depends on the margin of error you're willing to account for, and there are strict mathematical definitions for that. Gallup usually goes for a 3-4% margin of error - meaning a sample size of typically about 1150-1200.

    I believe they surveyed over 9,000 people in this study. And the samples were random. That would leave a miniscule margin of error.

    Now, you can complain all you want that Gallup did not cover *every* Islamic nation, and that it did not cover *every* Islamic culture (they are of course all different), but Gallup hit on every *major* relevant subset as pertains to our interests.

    Let's face it: while Indonesia has a large population, it is not particularly relevant to the War on Terror. At least at this point.

    But whatever. I realize that many of you would love to believe that the Islamic World does not hate us, and you're willing to hop onto any theory that will allow you to do so. Deceive yourself if that is your choice, I really don't care. But the Gallup results were valid. Blame the reporters if you just have to throw mud somewhere.

    Ever heard the saying "Statistics don't lie, people do"? It's true.
     
  2. Major

    Major Member

    Joined:
    Jun 28, 1999
    Messages:
    41,685
    Likes Received:
    16,213
    But whatever. I realize that many of you would love to believe that the Islamic World does not hate us, and you're willing to hop onto any theory that will allow you to do so.

    LOL... You're the one who insists on believing the opposite, no matter what the truth may be. The methods used is this poll were NOT scientifically accurate, by any stretch. To claim that the "Islamic World" hates us is completely and totally false according to the *real* results of the poll.

    In any poll, to get an accurate report, the poll results have to be extrapolated to the entire population they represent. Here, they weren't (at least not in the reported version). The 1000 or whatever people surveyed in Indonesia were NOT extrapolated to represent the entire 200 million population there vs the 1000 or whatever people surveyed in another country. In the reported results (which is all anyone cares about, because that's what people are arguing was messed up), that wasn't considered.

    As far as the reporting, Gallup reported it this way, so yes, it IS their fault.
     
  3. Jeff

    Jeff Clutch Crew

    Joined:
    Feb 14, 1999
    Messages:
    22,412
    Likes Received:
    362
    I'm glad treeman was here to clear everything up for us idiots who can't figure out how to read polling numbers. :rolleyes:

    Outstanding post, Major.
     
  4. treeman

    treeman Member

    Joined:
    Nov 27, 1999
    Messages:
    7,146
    Likes Received:
    261
    Major:

    Uh... Did you even bother checking out the polling methods? I gave you a link to it (the first one). Why don't you try reading about and understanding their methods instead of simply assuming that they're biased/invalidated?

    Gallup polled on a nation-by-nation basis, and only analyzed the results on a nation-by-nation basis. It was reported by CNN incorrectly - CNN applied the results from Kuwait and Indonesia as an amalgamation, Gallup did not. Just click on the damn link if you don't believe me.

    Jeff:

    I can tell you didn't bother checking out the methods either. Way to prove me right. :rolleyes:
     
  5. rimbaud

    rimbaud Member
    Supporting Member

    Joined:
    Nov 3, 1999
    Messages:
    8,169
    Likes Received:
    676
    tree, I think people were just going off of this:

    So where did reporters get the bogus aggregate numbers?
    From the Gallup Organization, said Andrea Stone, the reporter who wrote USA Today's page 1 story. The aggregates were clearly noted in the first fax sent by Gallup to USA Today and CNN, which also included the country-by-country results. "I didn't do the arithmetic," Stone said.
    Frank Newport, editor-in-chief of the Gallup Poll, acknowledged that the aggregates were included in the material supplied to news organizations. So that means Gallup itself is partially to blame for the snafu, right?
    "I don't know," he said.
    Newport said Gallup analysts repeatedly cautioned against using the aggregated numbers when the results were released at a seminar in Washington. But no such warning appeared on Gallup's first release to Stone or to reporters at CNN.


    It seems they provided two sets of results.
     
  6. Major

    Major Member

    Joined:
    Jun 28, 1999
    Messages:
    41,685
    Likes Received:
    16,213
    Gallup polled on a nation-by-nation basis, and only analyzed the results on a nation-by-nation basis. It was reported by CNN incorrectly - CNN applied the results from Kuwait and Indonesia as an amalgamation, Gallup did not.

    I know how they did the polling itself -- there's nothing wrong with that. However, the above is simply NOT TRUE. CNN, when they initially report the polling results, has a Gallup representative on the air explaining the results. HE stated the results that way, and then CNN repeated it over and over.

    I know the polling method, and I also know that Gallup consolidated the results and reported in this wildly incorrect way. Here are some quotes, in case you didn't read the original:

    If the results of the two countries were averaged together, which is what NCPP said Gallup did, the result suggests that about 20 percent of these Muslims seemed to view the attack as justified.

    So where did reporters get the bogus aggregate numbers?
    From the Gallup Organization, said Andrea Stone, the reporter who wrote USA Today's page 1 story. The aggregates were clearly noted in the first fax sent by Gallup to USA Today and CNN, which also included the country-by-country results.


    Frank Newport, editor-in-chief of the Gallup Poll, acknowledged that the aggregates were included in the material supplied to news organizations.

    Gallup should have never aggregated the stats like that, because its obvious that reporters will report the simplest numbers. Regardless of who's fault it was, the whole point of this thread is irresponsible reporting -- whether it was by CNN, USA Today, Gallup, or whomever -- and the lack of effort made in correcting said mistake.
     
  7. treeman

    treeman Member

    Joined:
    Nov 27, 1999
    Messages:
    7,146
    Likes Received:
    261
    Major:

    Well, I don't know what was said on TV when the poll first came out (I don't watch much TV), but I do know what Gallup says on its website, and the polls are not aggregated there.

    Again, bad reporting does not invalidate the results of the poll, as so many here appear to believe - that's my only real beef with this issue. Jeff saying the poll was "wrong", you saying it was "a badly done poll all around", etc - that is attacking the poll itself, not the reporting. The poll itself was solid, and the results still stand.

    If all anyone here is concerned about is the reporting of the results, then why didn't anyone mention it?
     
  8. Major

    Major Member

    Joined:
    Jun 28, 1999
    Messages:
    41,685
    Likes Received:
    16,213
    Jeff saying the poll was "wrong", you saying it was "a badly done poll all around", etc - that is attacking the poll itself, not the reporting.

    OK, I was referring to the reporting of it there. I was also responding to Azadre's issues that it claimed to represent the Muslim world when it really didn't. In that respect -- if the poll made that claim in how it was presented by Gallup -- it was badly done (if it was to respresent the Muslim world, then it was badly polled; if it wasn't, then it was badly presented).

    Regardless, my issue is that Gallup reported it to media outlets basically as follows:

    (simplified)
    Indonesia - 4% hate the US
    Kuwait - 36% hate the US
    etc
    Total - 20% hate the US

    Or whatever the final numbers were. The Totals line should have NEVER been tabulated that way by anyone. Gallup made a huge f*** up that got spread through the media, and there's been no attempt whatsoever to correct it outside of this little online article that haven found. That's just irresponsible journalism because the initial report is what everyone is going to remember.
     
  9. treeman

    treeman Member

    Joined:
    Nov 27, 1999
    Messages:
    7,146
    Likes Received:
    261
    From the Gallup polling methods page:

    Reporting

    The results of this project are for the most part discussed on a country-by-country basis without specific reference to an overall total. In certain graphic representations the "Total" bar is based on an unweighted sample of all interviews conducted and should be used for illustrative purposes only. Since the nine countries included in this sample are not intended to represent all predominantly Muslim countries across the world, since the residents in these nine countries do not represent all Muslims, and since there are significant variations among the nine countries in the responses to the questions included, we -- as noted above -- encourage readers using these data to examine them primarily on a country-by-country basis.


    Pretty self-explanatory, if you ask me.
     
  10. Major

    Major Member

    Joined:
    Jun 28, 1999
    Messages:
    41,685
    Likes Received:
    16,213
    From the Gallup polling methods page:

    That's fine and all, but if Gallup then turns around and does tabulate them and present the results as a sum-total (which they did), and they DO present the poll as a "Views of the Islamic World" (I don't know if they did this, but every news outlet used the exact same wording...) in their press releases for the poll, this little disclaimer tidbit doesn't really mean a whole lot. It's still irresponsible.

    Maybe CNN should have done more analysis, but if its presented by Gallup that way, I think the news media has every reason to assume they've presented it in a reasonably accurate way. After all, Gallup is the expert on polling and poll analysis, not CNN.

    "Total" bar is based on an unweighted sample of all interviews conducted and should be used for illustrative purposes only.

    On a side not, here's another problem ... it SHOULDN'T be used for illustrative purposes ... it shouldn't be used for ANY purpose because it's meaningless. It means just as much as if I said "60% of Texans voted Bush, and 50% of Floridians voted Bush, so 110% of Americans voted Bush" or something ridiculous like that. It's not a "total" in any sense of the word.
     
  11. haven

    haven Member

    Joined:
    Oct 22, 1999
    Messages:
    7,945
    Likes Received:
    14
    Treeman, you're a fool who has trouble with reading comprehension. All of your concerns are addressed in the original article I posted. You had nothing to say that wasn't said in the original article. Idiot.

    Here, why don't you read again...

    There you are! Your concerns addressed! It seems like you just can't pass up an alternative to attack anything or anyone that suggests that Arabs might not be terrible, after all...
     
    #31 haven, Mar 26, 2002
    Last edited: Mar 26, 2002
  12. Rocketman95

    Rocketman95 Hangout Boy

    Joined:
    Feb 15, 1999
    Messages:
    48,984
    Likes Received:
    1,445
    OK, I'm gonna poll 1000 basketball fans in Utah and 1000 baskeball fans in Texas, and I'll will only ask one question.

    How many of you like the Utah Jazz.

    I hypothesize that 98% of the Utah respondents will answer yes and that 4% of the Texans will answer no.

    Therefore, 51% of all basketball fans like the Jazz.
     
  13. treeman

    treeman Member

    Joined:
    Nov 27, 1999
    Messages:
    7,146
    Likes Received:
    261
    haven:

    Real constructive. Is this your logical mind at work, or are you just sloppy drunk?

    Needless to say, your post requires a :rolleyes: response. But...

    My *concern* here is that because this poll was initially reported incorrectly, you dimwits will simply (and I do mean "simply") think that the results of the poll are not valid. My argument here is that the results are still valid - are you arguing that?

    The way the results were presented was flawed, the research and results were not. What the reporting should have said is something along the lines of "In those countries polled, we are much disliked". I understand if you do not get the statistical principles at work here - you're training to become a liar - er, um, lawyer - right? I guess analyzing polling methods isn't your cup of tea. That's fine, and to each his own.

    My concern is valid I think, because we have already seen people jumping on here and insinuating that the poll is garbage. We have people here (you included) who are willing to overlook any evidence that we are widely disliked in the Islamic world, simply because that evidence was presented in a (somewhat) misleading way.

    The numbers are there, and they are unarguable. You are not even bothering to interpret the results - you just find a bunk story to discredit the presentation, and you're all set to maintain your fantasy of a benevolent and adoring Islamic world! You conveniently ignore the actual results of the survey, so that you can stick to your dream interpretation of how the world works.

    You are free to think what you want. You are also free to ignore any compelling evidence that contradicts what you want to think. That is your problem. As I've said many times here, what you think is irrelevant anyway...

    Have fun living in Fantasyland. Just know that while you stew in your self-imposed cauldron of ignorance, real people who have an actual understanding of what is really going on will act to defend your sorry ass from the people you want to believe like you. :)
     
  14. Major

    Major Member

    Joined:
    Jun 28, 1999
    Messages:
    41,685
    Likes Received:
    16,213
    The way the results were presented was flawed, the research and results were not. What the reporting should have said is something along the lines of "In those countries polled, we are much disliked".

    But this isn't true, either. The real result of the poll is that very few of the total population in all the countries polled dislike us. Sure, some countries have real issues with us, but in total (even if you assume these 9 countries are the "Islamic World" that they refer to), that's simply not true.

    For example, just using the Kuwait & Indonesia numbers, only about 4.5% of the people really dislike us (they would have reported 20% in this scenario). I don't know what the sum total is, but it is nowhere near the 53% of the people in these countries (as was initially reported). Since I'm not a Gallup member, I don't have access to the actual numbers. But given the countries surveyed, I don't think the results are above 10-20% in total. That is the problem with the way it was reported.
     
  15. Jeff

    Jeff Clutch Crew

    Joined:
    Feb 14, 1999
    Messages:
    22,412
    Likes Received:
    362
    haven: Don't make it personal, man, no matter how much it bugs you. It just complicates things on here and pisses people off.

    treeman: Let me just say thanks for saying that because I thought I had read all the narrow-minded, condescending things people could possibly post in my 6 years here and then you go and post this. I can die a happy man now because I've seen everything. :)
     
  16. treeman

    treeman Member

    Joined:
    Nov 27, 1999
    Messages:
    7,146
    Likes Received:
    261
    Major:

    No, the real result of the poll was that in the countries polled, there is a very strong anti-US sentiment - it is far more widespread than many here want to believe. I do not see how you can get "very few of them dislike us" out of those results, unless you're simply in denial.

    The quibble here is over the "total" tally, and that was flawed, as there are significant differences among the various countries polled. Quite frankly, it is hard not to notice that the numbers regarding the "dislike" issue are much higher in the Arab countries polled than in the non-Arab countries... Indonesia is relatively indifferent towards us, while the Saudis and Jordanians hate our guts with a passion.

    What are you talking about? In Indonesia, about 4% of them dislike us, but in Kuwait about 36% of them dislike us. Remember that whole 'country-by-country' thing? Your problem is supposed to be with the aggregate totaling of the poll, and yet here you are doing exactly what you claim to dispute - only you're trying to artificially push the numbers down to the Indonesia level! Talk about denial...

    I do realize that this is a pointless discussion, as you are going to believe what you want to believe and use any excuse to try and rationalize the results as invalid. But I just can't let details like that slide... ;)

    Jeff:

    We'll be defending your misguided butt too. :) I do not expect a thank-you, either - just don't spit on me when I get off the plane in as few years...
     
  17. Major

    Major Member

    Joined:
    Jun 28, 1999
    Messages:
    41,685
    Likes Received:
    16,213
    What are you talking about? In Indonesia, about 4% of them dislike us, but in Kuwait about 36% of them dislike us. Remember that whole 'country-by-country' thing? Your problem is supposed to be with the aggregate totaling of the poll, and yet here you are doing exactly what you claim to dispute - only you're trying to artificially push the numbers down to the Indonesia level! Talk about denial...

    What are YOU talking about?

    Let's see.. let's take the issue of Affirmative Action and assume 10% of America is Black , 90% are White. Imaginary poll:

    I poll 100 Black people and 100 White people:

    100 Black people - 90% agree with affirmative action
    100 White people - 20% agree with affirmative action

    Now, the way Gallup reported it, they would have said 55% of Americans agree with AA, which is completely and totally false. The proper result is that you extrapolate each significant group separately... the result is 20% x 90% + 90% x 10% of Americans agree, which is 27% of people. That is the scientifically accurate result.

    Just taking Kuwait and Indonesia, the inaccurate result is 20%. The accurate result is 4.6%. This is not artifically created or improper in any way. It's the scientifically proper way to analyze the poll. 4.6% of the Muslims in these two countries dislike the US. Those are simply the facts from their own poll. If you provide me the numbers for the other 7 countries, I can give you the accurate total results. Unfortunately, the Gallup site requires me to pay $1,250 for those numbers, and that really isn't worth it for me.
     
  18. Rocketman95

    Rocketman95 Hangout Boy

    Joined:
    Feb 15, 1999
    Messages:
    48,984
    Likes Received:
    1,445
    I still don't understand how anyone could see that the actual results of the poll where not what Gallup or CNN stated they were.

    Maybe I'm a stupid liberal who's in denial, but I really don't understand.

    So, again, if I ask 1000 basketball fans in Utah and 1000 basketball fans in Texas who their favorite team was, and all 980 residents of Utah answered the Jazz, and 40 people in Texas answered the Jazz, you would take that to mean that 51% of all basketball fans in the two states loved the Jazz, regardless of the number of basketball fans there were in each state?
     
  19. treeman

    treeman Member

    Joined:
    Nov 27, 1999
    Messages:
    7,146
    Likes Received:
    261
    No, the result for Indonesia is around 4%, and the result for Kuwait is around 36%. Where in the hell do you come up with 4.6% from that?

    And again, you're aggregating - and I frankly do not see how you're arriving at that number at any rate, even considering an aggregate, as the sample was weighted in the first place (you're not trying to proportionalize the actual populations, are you? Wrong). I am trying to tell you here that aggregation is not the way this should be done - supposedly your whole argument here.

    And BTW, have you totally forgotten about the "Reporting" quote I posted? "Scientifically proper"... S*it, even Gallup is advising against it.

    The results are still valid on a country-by-country basis, no matter how you try to stack it up.
     
  20. Rocketman95

    Rocketman95 Hangout Boy

    Joined:
    Feb 15, 1999
    Messages:
    48,984
    Likes Received:
    1,445
    OK, so what you have is the country with the most Muslims isn't that hostile towards Americans, while a tiny pissant country that we saved from Sadaam is.
     

Share This Page