Statistics Done Wrong – The woefully complete guide
1–10 of 71 posts
Re: Statistics Done Wrong – The woefully complete guide
#2I like that he references Huff's "How to lie with statistics" in the first sentence of the intro. That was the book that came to mind when I saw the subject. Also reminds me of the Twain quote, "There are three types of lies: Lies, Damned Lies, and Statistics."
But despite this, statistics done well are very powerful.
Re: Statistics Done Wrong – The woefully complete guide
#3As a graduate student in the life sciences, I was required to take a course on ethical conduct of science. This gave me the tools to find ethical solutions to complex issues like advisor relations, plagiarism, authorship, etc. We were also taught to keep good notes and use ethical data management practices - don't throw out data, use the proper tests, etc. Unfortunately, we weren't really taught how to do statistics "the right way." It seems like this is equally important to ethical conduct of science. Ignorance is no excuse for using bad statistical practices - it's still unethical. By the way, this is at (what is considered to be) one of the best academic institutions in the world.
Re: Statistics Done Wrong – The woefully complete guide
#4I like that he references Huff's "How to lie with statistics" in the first sentence of the intro. That was the book that came to mind when I saw the subject. Also reminds me of the Twain quote, "There are three types of lies: Lies, Damned Lies, and Statistics." But despite this, statistics done well are very powerful.
With respect to that Twain/Disraeli quote, my friend who is a professor of statistics tells me that he cannot go to a party and say what he does for a living without someone repeating it smirkingly.
Re: Statistics Done Wrong – The woefully complete guide
#5One thing that constantly saddens me about statistics is that a large amount of energy is expended using is almost correctly to "prove" something that was already the gut feel. Even unbiased practitioners can be lead astray [1] but standards on how not to intentionally lie with statistics are very useful.
[1] http://euri.ca/2012/youre-probably-polluting-your-statistics...
Re: Statistics Done Wrong – The woefully complete guide
#6One thing that constantly saddens me about statistics is that a large amount of energy is expended using is almost correctly to "prove" something that was already the gut feel. Even unbiased practitioners can be lead astray [1] but standards on how not to intentionally lie with statistics are very useful. [1] http://euri.ca/2012/youre-probably-polluting-your-statistics...
There's no way to tell whether or not that "gut feel" is accurate without proof. Often it's right, but occasionally it's very, very wrong (cancer risk and Bayes theory provides a good illustration: http://betterexplained.com/articles/an-intuitive-and-short-e...). Consequently it's still worthwhile proving things even when they're seemly obvious.
Re: Statistics Done Wrong – The woefully complete guide
#7The greatest problem of statistical analysis is throwing out observations which do not fit the bill. All analysis should be thoroughly documented with postmortems.
Re: Statistics Done Wrong – The woefully complete guide
#8One thing that constantly saddens me about statistics is that a large amount of energy is expended using is almost correctly to "prove" something that was already the gut feel. Even unbiased practitioners can be lead astray [1] but standards on how not to intentionally lie with statistics are very useful. [1] http://euri.ca/2012/youre-probably-polluting-your-statistics...
There's no way to tell whether or not that "gut feel" is accurate without proof. Often it's right, but occasionally it's very, very wrong (cancer risk and Bayes theory provides a good illustration: http://betterexplained.com/articles/an-intuitive-and-short-e... ). Consequently it's still worthwhile proving things even when they're seemly obvious.
"Consequently it's still worthwhile proving things even when they're seemly obvious."
Not true. It's all about risk vs. payoff. Some things are low risk enough we can go by gut, others we need more evidence. It's all about tuning for false positives and negatives.
EDIT: Added quote of what I was responding to.
Re: Statistics Done Wrong – The woefully complete guide
#9I like that he references Huff's "How to lie with statistics" in the first sentence of the intro. That was the book that came to mind when I saw the subject. Also reminds me of the Twain quote, "There are three types of lies: Lies, Damned Lies, and Statistics." But despite this, statistics done well are very powerful.
With respect to that Twain/Disraeli quote, my friend who is a professor of statistics tells me that he cannot go to a party and say what he does for a living without someone repeating it smirkingly.
Isn't that why the name "Data Science" was invented?
Re: Statistics Done Wrong – The woefully complete guide
#10One thing that constantly saddens me about statistics is that a large amount of energy is expended using is almost correctly to "prove" something that was already the gut feel. Even unbiased practitioners can be lead astray [1] but standards on how not to intentionally lie with statistics are very useful. [1] http://euri.ca/2012/youre-probably-polluting-your-statistics...
There's no way to tell whether or not that "gut feel" is accurate without proof. Often it's right, but occasionally it's very, very wrong (cancer risk and Bayes theory provides a good illustration: http://betterexplained.com/articles/an-intuitive-and-short-e... ). Consequently it's still worthwhile proving things even when they're seemly obvious.
He said he's saddened by people using statistics "almost" correctly to prove what was already "gut feel". I can't quite tell whether his real concern is what you thought he was saying (i.e., wasted effort), or whether it's the "almost" (but not quite correctly) part, i.e., that researchers use statistics to wrongly prove the gut feel. If it's the latter, then I think a big part of the problem is that people don't understand statistics well enough, not that they intentionally misuse it.