Register | Sign In


Understanding through Discussion


EvC Forum active members: 65 (9162 total)
2 online now:
Newest Member: popoi
Post Volume: Total: 915,806 Year: 3,063/9,624 Month: 908/1,588 Week: 91/223 Day: 2/17 Hour: 0/0


Thread  Details

Email This Thread
Newer Topic | Older Topic
  
Author Topic:   Darwinism Cannot Explain The Peacock
Dr Adequate
Member (Idle past 284 days)
Posts: 16113
Joined: 07-20-2006


(3)
Message 91 of 165 (689322)
01-29-2013 3:14 PM
Reply to: Message 80 by Arriba
01-29-2013 11:19 AM


And May God Have Mercy On Your Soul
So my message to any other users in this forum who might have suggested that I had "[constructed] a rather elaborate fantasy to explain how the data we have might support the sexual selection hypothesis, and the hypothesis might still be wrong..." is simple: Take a remedial math class and get back to me.
It was I who suggested that. I have a Ph.D. in math, and am profoundly unimpressed by your silly games with Bayesian statistics, and by the silly mistakes you've made while playing them.
Let's look at what you did wrong.
(1) Petrie's study does not investigate whether sexual selection can explain the tail of the peacock. It investigates whether and to what degree peahens select peacocks according to the qualities of their tails. These questions are not co-extensive, and so when you're assigning figures to P(H) and P(E|~H) the two Hs don't actually stand for the same thing.
(2) The figure you've used for P(H) is merely a quantification of your prejudices, i.e. you pulled it out of your ass. You can't really improve a statistical analysis by adding in a datum you've made up based on your personal bias. This is why the nature and magnitude of your biases are not taught in "remedial math class".
As with your previous post, what you've hit on is a method where you can deny a fact, no matter how true and no matter how well-supported by statistical evidence --- just by choosing P(H) to be low enough. If, for example, you want to deny that smoking causes cancer (as you might if you worked for a tobacco company) all you have to do is declare that it is "a priori" unlikely, assign it a small enough value, and all the evidence we have won't push P(H|E) over 50%. And then you can point out that P(H|E) is less than 50%, and that you've proved this with MATH!!! Oh, and by taking your prejudices as data.
One has to be very careful using Bayesian math in the context that P(H) is a guess rather than a known frequency. You have exercised no such caution, and this makes your whole number-juggling worthless. What you have proved is simply that if you are sufficiently biased against a proposition, it is hard for statistical evidence to make you believe it. Well, we knew that.
(3) And now it's time for you to pull another figure out of your ass --- the figure for P(E|~H). You don't actually know the p-value reported in Petrie's paper, as you admit. So you make it up --- and you choose the largest value consistent with the paper being published at all.
This breaks one of the cardinal rules of argument: that if you are trying to prove something, and you don't know a figure, it is illegitimate to assume the figure most favorable to your argument. To construct a strong argument, you use the figure least favorable to your argument; under certain circumstances it might be acceptable to use a figure based on an average.
For example, if I wish to prove that you will never become heavyweight champion of the world, and I don't know your height, it would be illegitimate for me to assume that you are as short as is consistent with you being a human being. On the contrary, since I'm trying to prove it impossible, I'd have to assume that you were the ideal height to become heavyweight champion of the world.
(If I merely wished to prove it unlikely, I might be excused for assuming that you were of average height.)
(4) And then you pull a third figure out of your ass when you claim that P(E|H) is 0.95.
I've saved the best 'til last.
Would you like to tell the class where you got that figure, Arriba?
Actually, I think the class can guess. You subtracted your (made-up) figure for P(E|~H) from 1, didn't you?
Really, Arriba?
And then you think yourself fit to lecture people on probability theory and to tell people other than yourself that they should "take a remedial math class". When that little maneuver tells us all we need to know about your mathematical literacy and grasp of probability theory.
Edited by Dr Adequate, : No reason given.
Edited by Dr Adequate, : No reason given.

This message is a reply to:
 Message 80 by Arriba, posted 01-29-2013 11:19 AM Arriba has replied

Replies to this message:
 Message 103 by Arriba, posted 02-01-2013 9:45 AM Dr Adequate has replied

  
subbie
Member (Idle past 1254 days)
Posts: 3509
Joined: 02-26-2006


(2)
Message 92 of 165 (689331)
01-29-2013 4:04 PM
Reply to: Message 83 by Arriba
01-29-2013 12:43 PM


Re: Unprovable Postulates
Although you may find it strange, I don't see any difference between their position and yours. You are two sides of the same coin - a bunch of gibbering idiots insisting that the truths printed in your holy books are beyond questioning.
Exactly right!
Oh, except for the fact that there is no holy book of science that contains truths that are beyond questioning. In fact, it's just the opposite; everything is subject to questioning in science, nothing is gospel and everything is held tentatively, pending the discovery of new evidence or a better explanation for the evidence we have.
Otherwise, you've hit the nail on the head. Well done.

Ridicule is the only weapon which can be used against unintelligible propositions. -- Thomas Jefferson
We see monsters where science shows us windmills. -- Phat
It has always struck me as odd that fundies devote so much time and effort into trying to find a naturalistic explanation for their mythical flood, while looking for magical explanations for things that actually happened. -- Dr. Adequate
Howling about evidence is a conversation stopper, and it never stops to think if the claim could possibly be true -- foreveryoung

This message is a reply to:
 Message 83 by Arriba, posted 01-29-2013 12:43 PM Arriba has not replied

  
Dr Adequate
Member (Idle past 284 days)
Posts: 16113
Joined: 07-20-2006


(1)
Message 93 of 165 (689358)
01-30-2013 1:42 AM
Reply to: Message 83 by Arriba
01-29-2013 12:43 PM


Re: Unprovable Postulates
I can't count the number of times I've heard that science has put a man on the moon and how the Internet wouldn't be possible without science. People roll their eyes and say how hard they laugh that someone uses their computer to say that science doesn't work.
On the other hand, I can't count the number of times I've heard that the Earth wouldn't exist without Jesus Christ, much less the Sun, and my body, and there would be no plants or animals to eat. They say that they laugh every time someone stands on the Earth and says they don't believe in God.
The difference would be that the first set of propositions is known to be true.
Here's a news flash for you - the most important reasons I have the computer I'm typing on are three:
1. Double-entry accounting.
2. Charles Babbage, mathematician, and his differential engine.
3. Six Sigma Statistical methods.
Where does electricity come on your list?

This message is a reply to:
 Message 83 by Arriba, posted 01-29-2013 12:43 PM Arriba has replied

Replies to this message:
 Message 106 by Arriba, posted 02-01-2013 9:58 AM Dr Adequate has replied

  
Dr Adequate
Member (Idle past 284 days)
Posts: 16113
Joined: 07-20-2006


(4)
Message 94 of 165 (689374)
01-30-2013 4:05 AM
Reply to: Message 80 by Arriba
01-29-2013 11:19 AM


Kicking The Ball Through The Goalposts
Researchers seeking to alleviate their ignorance - nonsense! They're only interested in getting funding, ideally federally funding, and that funding is intended to aid them in procreating as best as they can in order to further spread their genes. Truth has nothing to do with it so stop pretending. Science is how people who can't kick the ball through the goalposts try to get laid.
Even given that dubious proposition, it is obvious that scientists would still try to find the truth, because that is in fact how they gain their prestige. If their rockets fell over on the launching pads, if their vaccines made people more likely to contract diseases, if the nuclear reactors they designed took in more electricity than they put out, if their predictions of eclipses were no better than chance ... etc, etc ... then they would be objects of derision and scorn, and would not get the hot women. Or the funding. Let their motives be as venal as you choose to imagine, their venal motives would be a motive to be right.
In the same way, someone who tried to get money and women by "kicking the ball through the goalposts" would have a powerful incentive to kick the ball through the goal posts rather than kicking it somewhere else.
Edited by Dr Adequate, : No reason given.

This message is a reply to:
 Message 80 by Arriba, posted 01-29-2013 11:19 AM Arriba has not replied

Replies to this message:
 Message 95 by Pressie, posted 01-30-2013 4:41 AM Dr Adequate has not replied

  
Pressie
Member
Posts: 2103
From: Pretoria, SA
Joined: 06-18-2010


Message 95 of 165 (689375)
01-30-2013 4:41 AM
Reply to: Message 94 by Dr Adequate
01-30-2013 4:05 AM


Re: Kicking The Ball Through The Goalposts
Hey, you forgot the most important for me; if mining companies spend billions of dollars, then not finding the gold all those geologists indicated would be there, boy, would those geologists be in real trouble...geology works.

This message is a reply to:
 Message 94 by Dr Adequate, posted 01-30-2013 4:05 AM Dr Adequate has not replied

  
Bolder-dash
Member (Idle past 3629 days)
Posts: 983
From: China
Joined: 11-14-2009


Message 96 of 165 (689400)
01-30-2013 11:15 AM
Reply to: Message 79 by Blue Jay
01-29-2013 11:11 AM


Female chimpanzees have beards in about the same way that Richard Nixon had a sense of humor. To claim that they have beards in any way close to what male humans have is just downright funny. Quit monkeying around.
But there is obviously a larger point. Traits in your world come about by accident and then spread simply as a result of better survival. So when we talk about traits like beards, we must find a reason that they would make better survival, if your theory is to make sense. So you can try to think of reasons why they would give any significant advantage,but it should make sense. The sense that RAZD says it makes,is that if a female doesn't have a beard she would look adolescent like, as if looking adolescent like would help one to get mates easier. Not only does this not happen in the ape world-that adolescent looking apes mate easier, but it doesn't even happen in the human world-as why should it.
You situation becomes even trickier when you start creating stories,which even CONTRADICT earlier stories you created to explain breasts and hips.
But then your situation becomes absolutely nonsensical when you start claiming that traits like a peacocks tail are advantageous just because they are descibed as being advantageous...i.e. They look good and therefore they must be. Or that they are so hard to live with, that one must be healthy to survive with it-because its such a burden. Like if a man was born with a 200 pound lump on his back. If he could still survive with that,he must be very very healthy to carry that around all day.
I mean, Dr. A even went so far as to just say, woman without beards look good, so of course they mate easier. That you must admit is just patently stupid, in the context of your theory. Why would something be described and thought of as looking good? Even you must admit that is dumb, right?
Now you have a tail, which makes it harder to fly, which makes you slower, and more susceptible to being eaten, and yet somehow you have to be able to also describe this in a way that makes sense as an advantge. So your side is left bumbling with the notion that, well, they survived,so see that proves it must be.
Well, no it doesn't, because there is another option. And that is that traits exist that natural selection would never favor. Like hair and beards down to the floor. Life is not describable in a survival only paradign. Life is chock full of wonders which are cool,but which would never make the difference between life and death. Like an eyebrow. Its convenient, but its not going to insure your survival. Or a beard on a chimpanzee, because frankly chimpanzees don't give a **** who they ****... just like all other animals, they just aren't that choosey.
You don't only have an entire closet full of traits that you would have to rationalize to make your theory make sense,you have an entire planets worth of traits that need rationalizing. And when you rationalize them long enough, all the reasons you give start to contradict each other.
And so what does science do,because they already know theirtheory has to be right, they have already bet the house on it. Well,they just throw out the results which don't help them.or they just make up new stories and hope that everyone forgot the last story-just like RAZD has done. of course we want fully developed woman,er I mean adolescent woman. Of course we want the birds which fly fastest,er,I mean which are the most beautiful. Why are parrots beautiful, well,because look at them, they have survived! Don't you think they look good. Oh no wait, that's only for the male birds,because females care about looks most. No,sorry, that's wrong too,females want providers,males want the looks,thats why....
Wait, can I start again from the beginning.
If you can take any trait you want,and call it a survival advantge just because you need to,that is not much of a theory. And its not gish gallop either, just because you say it is.

This message is a reply to:
 Message 79 by Blue Jay, posted 01-29-2013 11:11 AM Blue Jay has replied

Replies to this message:
 Message 97 by NoNukes, posted 01-30-2013 11:57 AM Bolder-dash has not replied
 Message 98 by RAZD, posted 01-30-2013 12:09 PM Bolder-dash has not replied
 Message 99 by AZPaul3, posted 01-30-2013 1:28 PM Bolder-dash has not replied
 Message 100 by Blue Jay, posted 01-30-2013 4:08 PM Bolder-dash has not replied
 Message 102 by Dr Adequate, posted 01-30-2013 4:55 PM Bolder-dash has not replied

  
NoNukes
Inactive Member


(1)
Message 97 of 165 (689405)
01-30-2013 11:57 AM
Reply to: Message 96 by Bolder-dash
01-30-2013 11:15 AM


Female chimpanzees have beards in about the same way that Richard Nixon had a sense of humor.
They've got beards. Anyone can see them in the pictures. And we are not descended from chimps, anyways. Humans and chimps have a common ancestor.
Well, no it doesn't, because there is another option. And that is that traits exist that natural selection would never favor. Like hair and beards down to the floor.
Except that hardly any of us can grow hair and beards down to the floor. You are describing a trait that is relatively rare then and asking us why it has not been removed from the gene pool.
The selection pressure to remove long hair if there was such a thing, would have been ameliorated as soon as humans figured out how to cut or tie up their hair. My daughter has hair that is too long to play basketball with, so she binds it up before stepping onto the court.
Further, who cares? If the given explanation is wrong, then it is just wrong.
You situation becomes even trickier when you start creating stories,which even CONTRADICT earlier stories you created to explain breasts and hips.
I agree with you that some of the stories given here are ridiculous and that you are right to question them. But nobody here is doing any real research. People are just postulating natural selection and mutation based processes that might give the final result.

Under a government which imprisons any unjustly, the true place for a just man is also in prison. Thoreau: Civil Disobedience (1846)
The apathy of the people is enough to make every statue leap from its pedestal and hasten the resurrection of the dead. William Lloyd Garrison.
If there is no struggle, there is no progress. Those who profess to favor freedom, and deprecate agitation, are men who want crops without plowing up the ground, they want rain without thunder and lightning. Frederick Douglass

This message is a reply to:
 Message 96 by Bolder-dash, posted 01-30-2013 11:15 AM Bolder-dash has not replied

  
RAZD
Member (Idle past 1404 days)
Posts: 20714
From: the other end of the sidewalk
Joined: 03-14-2004


(1)
Message 98 of 165 (689406)
01-30-2013 12:09 PM
Reply to: Message 96 by Bolder-dash
01-30-2013 11:15 AM


sexual selection alive and well
Female chimpanzees have beards in about the same way that Richard Nixon had a sense of humor. To claim that they have beards in any way close to what male humans have is just downright funny. Quit monkeying around.
Female chimps have hairs on their chins. Young chimps do not appear to have hairs on their chins, but vellus hair is not easy to see, being very fine and relatively transparent.
Female humans have hairs on their chins -- it is just very fine vellus hair instead of terminal hair.
Male human have terminal hairs on their chins.
What has happened is that the progression of hair on female faces has been arrested during development at the vellus hair stage.
You situation becomes even trickier when you start creating stories,which even CONTRADICT earlier stories you created to explain breasts and hips.
Breasts and hips signal sexual readiness. The conflict between youthful appearance and sexual signal traits results in desire for a combination of traits that is not available ... and this has been ascertained by experiments. Such extreme shifting of sexual desired to outside the range of available alleles shows Fischerian Runaway Sexual Selection.
In populations where women have fat flat noses, the children also have fat flat noses.
The sale of razors to remove hair from armpits, legs etc. shows that selection for apparent bareness is alive and well in human society.
In cases where women have male pattern hair it is treated as a disability -- or as a circus performer.
Enjoy.

we are limited in our ability to understand
by our ability to understand
Rebel American Zen Deist
... to learn ... to think ... to live ... to laugh ...
to share.


Join the effort to solve medical problems, AIDS/HIV, Cancer and more with Team EvC! (click)

This message is a reply to:
 Message 96 by Bolder-dash, posted 01-30-2013 11:15 AM Bolder-dash has not replied

  
AZPaul3
Member
Posts: 8513
From: Phoenix
Joined: 11-06-2006
Member Rating: 5.2


(1)
Message 99 of 165 (689409)
01-30-2013 1:28 PM
Reply to: Message 96 by Bolder-dash
01-30-2013 11:15 AM


Traits in your world come about by accident and then spread simply as a result of better survival.
Says who? You?
Who says traits spread because of survival? And who cares since survival is not the key? What is the key? You still have not learned the concept.
So when we talk about traits like beards, we must find a reason that they would make better survival, if your theory is to make sense.
First, the theory will never make sense to you since you have no idea what the theory really says.
Second, why do we need to produce a reason for any and all specific traits?
If beards or peacock trains have any effect on (here's that word again) "fitness" (and that involves considerably more than mere survival) then it is in concert with all the other phenotypic traits. Can you grasp that simple concept? What level of effect does any specific trait have on the overall "fitness" of the individual? Can you say?
You cannot grasp the concept but you can grasp your strawman and think you know something. It is not bad enough to not know, you make no effort to know even when the concept is handed to you. You do not want to know.
So why are you in this discussion? To battle against something of which you are and want to remain totally ignorant? Rather foolish.

This message is a reply to:
 Message 96 by Bolder-dash, posted 01-30-2013 11:15 AM Bolder-dash has not replied

  
Blue Jay
Member (Idle past 2697 days)
Posts: 2843
From: You couldn't pronounce it with your mouthparts
Joined: 02-04-2008


(1)
Message 100 of 165 (689422)
01-30-2013 4:08 PM
Reply to: Message 96 by Bolder-dash
01-30-2013 11:15 AM


Hi, Bolder-dash.
bolder-dash writes:
Female chimpanzees have beards in about the same way that Richard Nixon had a sense of humor. To claim that they have beards in any way close to what male humans have is just downright funny. Quit monkeying around.
I'm not trying to monkey around: I'm trying to understand why photographic evidence that appears very clear to me does not seem clear to you.
Can you please describe exactly how chimpanzee beards differ from human beards? I honestly don't see what it is about their beards that makes you think they're so vastly different that they don't have any meaning to this discussion.
Also, I screwed up some "url" codes in my earlier post, but I have now fixed the problems that ruined my links to a female orangutan and a female gorilla. Orangutan beards are particularly interesting.
Bolder-dash writes:
Now you have a tail, which makes it harder to fly, which makes you slower, and more susceptible to being eaten, and yet somehow you have to be able to also describe this in a way that makes sense as an advantge. So your side is left bumbling with the notion that, well, they survived, so see that proves it must be.
I certainly can't deny the "bumbling" thing: we are, after all, relying on our meager human faculties to try to make sense of the wonders of the universe, so the fact that we suck at it isn't itself a very meaningful criticism. Science bumbles along like a toddler learning to walk; but, as I've been trying to explain to Arriba, there's really nothing for it: all we can do is keep bumbling around and do our best to learn from it as we go.
If you know of a better way to figure out what a peacock's feather-train is for, please enlighten us. And, if all you have is "turn to Jesus" or "read the Bible," then, surely one of you who has already turned to Jesus should already know what a peacock's feather-train is for. But, since neither of you is willing to answer this question, we can only assume that turning to Jesus hasn't magically imparted this knowledge on you, either. And so, we choose to continue bumbling, because, even though it's difficult and inefficient, bumbling can potentially get us a useful answer.
Bolder-dash writes:
Life is not describable in a survival only paradign.
I agree. As you rightly stated, the number of handicaps and hardships an organism can survive is very large, and there are a very large number of viable strategies for survival. But, as you have consistently failed to grasp, our paradigm is not "survival-only": our paradigm includes many separate processes of selection, some of which have very little, if anything, to do with "survival" at all.
Survival is a bare minimum standard of fitness. If you survive famines, parasites, predators, injuries, birth defects and natural disasters, but do not leave any offspring, your genes do not contribute to the future gene pool, and so, in evolutionary terms, you are a "failure," or a "dead end."
So, clearly, there should be selection for more than just basic "survival": there should be selection for things with little relevance to survival, but relevance to other metrics of success. There should be selection for anything that improves an organism's chances for successful reproduction. In fact, if the Theory of Evolution is correct, then successful reproduction, and not survival, is the "gold standard," so we should expect to see many examples of organisms whose drive to reproduce outweighs their drive to survive.
That's where we come up with the conundrum of the peacock. The feather-train is clearly a liability for the peacock, and surely it would be easier to survive without it. Why would an animal have such a trait; and how is it that they continue to thrive, despite having this unnecessary liability?
I would like to hear your ideas on this.
I think the feather-train somehow increases the peacock's chances of successful reproduction, even though I don't know how.
But, what do you think?
Do you think that peacocks simply suffer insignificant levels of predation, and thus the actual, realized liability is quite low?
Do you think God somehow designed the Indian ecosystem such that long feather-trains are not a liability?
-----
Bolder-dash writes:
You don't only have an entire closet full of traits that you would have to rationalize to make your theory make sense, you have an entire planets worth of traits that need rationalizing. And when you rationalize them long enough, all the reasons you give start to contradict each other.
I don't deny that it's a confused mess. But, given that there are no universal principles that force animals (or humans) to behave in clear, consistent, perfectly rational patterns, what else would you expect? Why would expect perfectly consistent behavior in such a complex system?
For example, for humans, women with large breasts and wide hips are considered attractive; but, a long, slender stomach is also considered attractive. So, "idealized" women in fiction frequently have both (e.g., Jessica Rabbit, Betty Boop, Barbie, etc.), which seems like an architecturally unsound arrangement.
How do you explain this sort of behavior, except under a context of inconsistent, contradictory signals?

-Bluejay (a.k.a. Mantis, Thylacosmilus)
Darwin loves you.

This message is a reply to:
 Message 96 by Bolder-dash, posted 01-30-2013 11:15 AM Bolder-dash has not replied

  
Taq
Member
Posts: 9970
Joined: 03-06-2009
Member Rating: 5.6


(3)
Message 101 of 165 (689425)
01-30-2013 4:16 PM
Reply to: Message 65 by Arriba
01-28-2013 1:08 PM


Re: The story is not complete.
Unfortunately, I couldn’t find the effect, he said. But the worst part was that when I submitted these null results I had difficulty getting them published. The journals only wanted confirming data.
This is actually one aspect of peer reviewed journals that many scientists have complained about. What we really need are journals that publish just negative results. Sadly, a lot of labs follow the same dead-ends because no one publishes failed hypotheses. There are a few papers that do get published based on negative results, but these are usually related to long standing hypotheses such as sexual selection in peacocks.
Personally I feel that this example is very relevant because it is about what we are talking about: Sexual selection and its supposed confirmation. Sure it gets confirmed - because of selection, reporting, and publishing bias. So what does that prove except that scientists are very good at finding the results they want/need to find? Little by little other studies that confirm it but with lesser and lesser strength will appear until the purported effect will only exist as a myth in the minds of militant evolutionary apologists in debate forums such as this one.
And yet you are able to point to published papers that challenge the perception that every feature of the peacock train is under sexual selection. So much for that bias.
Natural selection is, at best, a tautology. Anyone can state that the fittest will survive as long as you can define what "fit" means after you know who does and does not survive.
As algebra demonstrates, tautologies can still be true. Want to find out who owns the fastest car? Then allow them to race each other and see who wins. Natural selection is a fact in the same way that other tautological observations are facts.
The theory part is in determining if this observed mechanisms is responsible for the biodiversity we see today. To test this idea we compare genomes and look for patterns of selection between genes and different regions of the genome. This is not tautological. We can predict what natural selection should look like when we compare genomes, and we do find that signal.
I must say, though, I admire your faith that somehow somewhere science will complete the story. It's touching in its naive simplicity.
Your faith seems to be supported by the idea that science will not find an answer.

This message is a reply to:
 Message 65 by Arriba, posted 01-28-2013 1:08 PM Arriba has replied

Replies to this message:
 Message 104 by Arriba, posted 02-01-2013 9:47 AM Taq has replied

  
Dr Adequate
Member (Idle past 284 days)
Posts: 16113
Joined: 07-20-2006


(1)
Message 102 of 165 (689432)
01-30-2013 4:55 PM
Reply to: Message 96 by Bolder-dash
01-30-2013 11:15 AM


Now you have a tail, which makes it harder to fly, which makes you slower, and more susceptible to being eaten, and yet somehow you have to be able to also describe this in a way that makes sense as an advantge. So your side is left bumbling with the notion that, well, they survived,so see that proves it must be.
Oh, look, it's the Argument From Undesign.
Yeah, peacock tails are stupid, aren't they? Which proves that they must have been created by an all-wise god who also happens to be a total moron. The tail of the peacock is such a bad lousy idea that instead of being produced by a process which you guys habitually summarize as "random chance", it must have been designed by infallible omniscience ... on one of its off-days.

This message is a reply to:
 Message 96 by Bolder-dash, posted 01-30-2013 11:15 AM Bolder-dash has not replied

  
Arriba
Junior Member (Idle past 3610 days)
Posts: 22
From: Miraflores, Lima, Peru
Joined: 01-24-2013


Message 103 of 165 (689587)
02-01-2013 9:45 AM
Reply to: Message 91 by Dr Adequate
01-29-2013 3:14 PM


Re: And May God Have Mercy On Your Soul
Deja que los perros ladran - es seal que avanzamos.
"It can be proven that most claimed research findings are false," so says Ioannidis and continues, "...that the high rate of nonreplication (lack of confirmation) of research discoveries is a consequence of the convenient, yet ill-founded strategy of claiming conclusive research findings solely on the basis of a single study assessed by formal statistical significance, typically for a p-value less than 0.05." Why is that? Because the pre-study probability is normally quite low. His example:
Let us assume that a team of investigators performs a whole genome association study to test whether any of 100,000 gene polymorphisms are associated with susceptibility to schizophrenia. Based on what we know about the extent of heritability of the disease, it is reasonable to expect that probably around ten gene polymorphisms among those tested would be truly associated with schizophrenia, with relatively similar odds ratios around 1.3 for the ten or so polymorphisms and with a fairly similar power to identify any of them. Then R = 10/100,000 = 10−4, and the pre-study probability for any polymorphism to be associated with schizophrenia is also R/(R + 1) = 10−4. Let us also suppose that the study has 60% power to find an association with an odds ratio of 1.3 at α = 0.05. Then it can be estimated that if a statistically significant association is found with the p-value barely crossing the 0.05 threshold, the post-study probability that this is true increases about 12-fold compared with the pre-study probability, but it is still only 12 10−4.
Now let us suppose that the investigators manipulate their design, analyses, and reporting so as to make more relationships cross the p = 0.05 threshold even though this would not have been crossed with a perfectly adhered to design and analysis and with perfect comprehensive reporting of the results, strictly according to the original study plan. Such manipulation could be done, for example, with serendipitous inclusion or exclusion of certain patients or controls, post hoc subgroup analyses, investigation of genetic contrasts that were not originally specified, changes in the disease or control definitions, and various combinations of selective or distorted reporting of the results. Commercially available data mining packages actually are proud of their ability to yield statistically significant results through data dredging. In the presence of bias with u = 0.10, the post-study probability that a research finding is true is only 4.4 10−4. Furthermore, even in the absence of any bias, when ten independent research teams perform similar experiments around the world, if one of them finds a formally statistically significant association, the probability that the research finding is true is only 1.5 10−4, hardly any higher than the probability we had before any of this extensive research was undertaken!
------------------------
Accordingly his a priori odds are 10/100,000 or 0.0001 but when I start by granting you 0.01 - a number 100 times more generous than the one he uses you whine about it like a little bitch. You also didn't pick up, I note, that in the final calculation I gave you P(H|E) = 0.01 / 0.118 - that is, two figures, when the formula specifically has three variables. What you didn't notice is that I assumed P(E|H) was 1 during the last step, even though I had previously assumed it was only 0.95. This was my way of being generous to you - an extra 5 percent bonus so that no one could say I had been unduly parsimonious in my approach. You also seem to think I should grant you some sort of study odds better than p=0.05 despite the fact that the article I was quoting from uses that number specifically.
But we both know you didn't read that article. Why would you when you already know the truth as though it were revealed to you from on high? Your real goal is to stamp out the heresy you see before you. Like the rest of your fellow cult members, you have a mind like concrete: All mixed up and permanently set.
Really, though, my analysis was even more generous than the above makes out. As you would know if you had bothered to Google Petrie's study, her study design was simple. She found a group of peacocks, captured half of the males, and disfigured portions of their tail. Later she lurked in the bushes watching their copulation and drew her conclusions. Let's analyze this simple study design for the possible reasons why it worked:
Reason 1. Maybe she's right. Maybe peahens really do like cocks with lots of eyes.
Reason 2. She was biased and consciously or subconsciously created the result she sought.
Reason 3. The peacocks that had been mangled were traumatized and spent less time seeking out hens and more time hiding - especially when they saw their former assailant lurking in the bushes.
Reason 4. The peahens didn't care about the number of tail eyes, per se but discriminated against the mutilated cocks because of the mulilation alone.
Reason 5. The peahens had grown accustomed to the males having a certain number of eyes in their tails due to long associations whereas had they grown up around males with lower numbers of eyes they would have been accustomed to that and would have rejected males with "excessive" numbers of eyes as out of the norm.
Perhaps there are even more reasons that I just can't think of right now. Be that as it may a simple application of the principle of insufficient reason suggests that we assign a 20 percent probability to her claim in light of her study. Even if we assume that her study being replicated ruled out reason 2 (something I am prepared to dispute) we might be able to get the chance up to 0.25 but not more.
And that's before the study I previously mentioned that found no relationship. Dare I suggest that those numbers will decline sharply?
As for your claim that "The figure you've used for P(H) is merely a quantification of your prejudices..." my response is simple: Se cree ladrn que todos son de su condicin.

"...nobody to date has yet found a demarcation criterion according to which Darwin can be described as scientific..." - Imre Lakatos

This message is a reply to:
 Message 91 by Dr Adequate, posted 01-29-2013 3:14 PM Dr Adequate has replied

Replies to this message:
 Message 109 by Dr Adequate, posted 02-01-2013 11:28 AM Arriba has replied
 Message 110 by Taq, posted 02-01-2013 11:28 AM Arriba has not replied
 Message 116 by Blue Jay, posted 02-02-2013 3:09 AM Arriba has replied

  
Arriba
Junior Member (Idle past 3610 days)
Posts: 22
From: Miraflores, Lima, Peru
Joined: 01-24-2013


Message 104 of 165 (689588)
02-01-2013 9:47 AM
Reply to: Message 101 by Taq
01-30-2013 4:16 PM


Re: The story is not complete.
It's not that tautologies are often true - tautologies are always true they're just not very useful.
"Wherever you go, there you are." No one can say that's not true - it's just not a helpful observation.

"...nobody to date has yet found a demarcation criterion according to which Darwin can be described as scientific..." - Imre Lakatos

This message is a reply to:
 Message 101 by Taq, posted 01-30-2013 4:16 PM Taq has replied

Replies to this message:
 Message 108 by Taq, posted 02-01-2013 11:21 AM Arriba has not replied

  
Arriba
Junior Member (Idle past 3610 days)
Posts: 22
From: Miraflores, Lima, Peru
Joined: 01-24-2013


Message 105 of 165 (689590)
02-01-2013 9:51 AM
Reply to: Message 88 by AZPaul3
01-29-2013 2:12 PM


Re: The story is not complete.
You greatly mistake the situation. If you had bothered to read the post (or been capable of understanding it) you would realize that we are saying that only one person took the exam and got a positive result. We then took a calculator and worked out the likelihood that this positive result was false and got a number well over 90 percent. Then we concluded that a test that was 99.5 percent accurate was not a panacea to all problems involving false positives.
This situation is comparable to published studies because published studies are almost always studies in which a positive relationship is claimed to have been found.
Additionally, of course, there's no way to know that the test is 99.5 percent accurate. All we can say for sure is that the test agrees with the Western Blot test 99.5 percent of the time.

"...nobody to date has yet found a demarcation criterion according to which Darwin can be described as scientific..." - Imre Lakatos

This message is a reply to:
 Message 88 by AZPaul3, posted 01-29-2013 2:12 PM AZPaul3 has replied

Replies to this message:
 Message 111 by Taq, posted 02-01-2013 11:30 AM Arriba has replied
 Message 115 by AZPaul3, posted 02-01-2013 8:17 PM Arriba has not replied

  
Newer Topic | Older Topic
Jump to:


Copyright 2001-2023 by EvC Forum, All Rights Reserved

™ Version 4.2
Innovative software from Qwixotic © 2024