|
Thread: Laelth's Election Projection | This thread is pages long: 1 2 3 · «PREV |
|
sirzapdos
Promising
Famous Hero
Open the pod bay doors, Hal.
|
posted November 05, 2004 06:57 AM |
|
|
Even though I doubt this is 100% verifiable, check this out.
____________
So I try to live a complicated world...
|
|
Binabik
Responsible
Legendary Hero
|
posted November 05, 2004 11:46 PM |
|
|
Ah, come on Bush supporters, are you going to let this IQ test stand? I'm not a Bush supporter, but this is everyone's favorite game. It sounds like fun and I wanna play!
It's called the game of statistics. And one of the favorite ways to play this game is called correlation and causation. The basic rule of the game is to provide statistics showing a correlation between two things. Then you leave the "cause" for the correlation un-stated. The cause is simply implied. It's really a lot of fun!
Fact: 100% of the times the Boston Redsox have won the World Series, and there was a presidential candidate from Boston, that candidate lost the election.
Conclusion: Well, 100% is pretty significant statistically, so there must be something to this. Oh, I know, the American people thought one big win for Boston in a year was enough. So they voted against Kerry to spread the joy of winning around. Yea, that's it, it must be.
Ridiculous? No, not at all, it's exactly the same thing as the IQ test results. With my example, it's easy to call it ridiculous. But can anyone state for absolute fact that my conclusions are wrong? With the provided statistics, you can't draw a conclusion either proving or disproving me.
Ah, so we don't have enough data to draw a conclusion.
I remember in school learning something called the "scientific method". Part of that method was that, when doing an experiment, only one variable can be changed at a time. I learned that when there is more than one variable, the results are invalid. At most you can claim that either A or B or C is true, or maybe some combination of those.
The IQ results.
Since people have been using the words smart and stupid in recent posts, I think maybe the implied conclusion of this test is that Kerry supporters are smarter than Bush supporters. After all the statistics "proved" that was true didn't they?
But wait, maybe we don't have enough data. Maybe there are other variables involved. With the provided data, all we can do is guess what those variables might be.
Is the data completely bogus? Maybe, but my guess is no.
Do IQ tests actually show how "smart" a person is? I think it does, but only to a degree. It's pretty narrow in it's scope. Take the term "musical genius" for example. Does an IQ test measure a persons ability to write or play music? Or to paint, or to design beautiful landscaping? Are these types of things a form of intelligence? And even with the "academic" nature of the tests, could education level or other things be a factor? After all, education and intelligence are two different things.
More data.
I don't have the numbers here, but there is data about urban, suburban and rural voting patterns. Basically, urban areas voted heavily in favor of Kerry. Suburban and rural areas leaned more toward Bush. So where are the heavily populated urban areas concentrated? Wow, the east and west coast, with Chicago and Detroit in the middle. Those just happen to be the states where Kerry won.
So the urban areas voted heavily for Kerry. I wonder how the urban areas and inner city score on IQ tests? Again I don't have numbers, but my guess is urban areas score lower than suburban areas. Does this mean people in urban areas are stupid? No, you can't draw that conclusion either, because there are other factors involved.
Again, I don't have the numbers, but hey, with lack of complete cross sectional data, we're just guessing anyway right? So let's redraw the map highlighting urban, suburban and rural areas. Now I'm picturing this map with lower IQs in the urban areas. Wow, it shows that Kerry voters are ...... leaves implied conclusion un-stated.
The problem with statistics is that if someone cares to look, they most likely could find statistical data supporting literally ANY conclusion they wanted. But the "support" for that conclusion is only implied.
Using the scientific method, the implied conclusions of both Sirzapdos' link and my own example are completely invalid.
____________
|
|
bort
Honorable
Supreme Hero
Discarded foreskin of morality
|
posted November 06, 2004 12:15 AM |
|
|
Quote: So the urban areas voted heavily for Kerry. I wonder how the urban areas and inner city score on IQ tests? Again I don't have numbers, but my guess is urban areas score lower than suburban areas. Does this mean people in urban areas are stupid? No, you can't draw that conclusion either, because there are other factors involved.
Now, I don't think that there is a 1:1 correlation between IQ and Kerry support either, but it's one thing to take statistics, present them without explanation and imply conclusions, it's quite another to make up statistics ("my guess is urban areas score lower than suburban areas.") and imply conclusions. What is your reasoning for guessing that urban areas would have a lower IQ?
____________
Drive by posting.
|
|
csarmi
Supreme Hero
gets back
|
posted November 06, 2004 01:24 AM |
|
|
I've already told somewhere else: this IQ table can't be valid.
Way too much differences amongst the states and the lower ones are too low. A state with average IQ of 85? 85 is dangerously low!
|
|
Binabik
Responsible
Legendary Hero
|
posted November 06, 2004 04:05 AM |
|
|
Bort
I stated that we don't have full statistical data, so I could only "guess" at other possibilities. Whether my guess is right or not doesn't matter. When concluding causation through correlation, the only thing that matters is the possibility that other factors "might" exist. I simply gave a few possibilities to make a point.
OK, so I haven't answered your direct question as to my reasoning.
First, it's a guess, but I think it's probably true. I also admit it could be wrong. As for my reasoning behind it, let me first point out my careful wording that urban areas "score lower". This intentional wording leaves it open for other factors besides intelligence. One of those factors is education level in urban areas vs suburban areas. No matter what the test makers say, education level is a factor in IQ tests. Another point of evidence is the argument commonly heard from the black community. They argue that language and culture based questions are biased against blacks. I add to that my own conclusion that there is a cultural difference between inner city and suburban areas. So at least some of their complaints may apply to inner city people in general as well as the black community. Another of my "guesses" is that the IQ tests are largely designed by middle and upper class, highly educated people. Any cultural differences show through in the tests. This is only part of it, but yes, I do have reasoning behind my statement.
You say I'm making up statistics and implying conclusions. When I say "Wow, it shows that Kerry voters are ...... leaves implied conclusion un-stated" it's borderline sarcasm. To *NOT* draw that conclusion was my entire point.
____________
|
|
|
|