"During the past century, though, a mutant form of math has deflected science’s heart from the modes of calculation that had long served so faithfully. Science was seduced by statistics, the math rooted in the same principles that guarantee profits for Las Vegas casinos.
[..]
It’s science’s dirtiest secret: The “scientific method” of testing hypotheses by statistical analysis stands on a flimsy foundation."
Tom Siegfried, ScienceNews
I consider this article to be further evidence for my favorite theory. And I do not need the 'mutant form of math' a.k.a. statistical analysis to know it has to be true.
PS: I will also add German finance minister Wolfgang Schaeuble to my list, who wants to use the German secret service to find and track those who sell the Euro.
Blog Archive
4 comments:
Despite the sensational lead, the article looks pretty solid. I'd say the odds are good that statistics are misused a lot, especially in the medical literature. On the other hand, the odds are pretty good that I might be wrong since I was never good at statistics.
I think the article can be summarized as follows: "5 out of 10 people don't know how to do statistics. Therefore the odds are high that something is wrong with statistics."
Notice that this is Science News magazine not the National Enquirer.
But what can I say - I don't even know how to calculate a sample mean.
I like a lot statistics, been learning for the past 2 years, and I think the following:
1. Nobody understand what the coefficient means, like r square z and so on.
2. Nobody understands that most of the statistics analysis are "valid" if the distribution of what is being analized is a normal one, which it does not happen very offen in the data outside nature, which is the one we are studying the most lately.
3.Nobody gets that there is no method in statistics that refers to cause and effect, nothing, you can never prove anything regarding that. There are all just relations.
4.People do not get that when they do a statistical test and for example the test variable is in the correct range it does not exactly prove our theory is correct, indeed nothing is ever proven correct in statistics, it does not work like logic. It just means that there is a high chance of us being right. But is still not exatrlly the same as something that is true in logic or math.
>> I like a lot statistics
welcome to the club!
>> Nobody understand what the coefficient means
i think this guy does (hint: check out his notebooks)
Post a Comment