Mistakes, Damned Mistakes, and Statistics

I'm working through The Cartoon Guide to Statistics by Larry Gonick with the goal of then getting through a more comprehensive book, then taking a research methods class so that I can do my own sampling. I highly recommend Gonick's The Cartoon Guide as an introduction to the subject.

There are various instructional videos on You Tube regarding statistics, but the most interesting is one presented by Oxford mathematician Peter Donnelly on how juries (and the rest of us) are easily confused and swayed by unsound mathematical arguments involving randomness and uncertainty.



Rather horrifying thought, isn't it - being convicted of a crime you did not commit by unsound statistical inference!

However, before the criminal case, Donnelly references what is called the False Positive Paradox: how a test that is 99% accurate can result in so many false test results indicating that well people have a serious disease.

Gonick references the same paradox. Assume that one person in one thousand has this serious illness, the test is 99% accurate, and the probability of a false positive test result is only 2%. Then supposed that you test positive. What is the probability that you actually have this disease?

Using Bayes' Theorem, we ultimately calculate that the percentage is .0472; that is, that out of all those who test positive for the disease, less than 5% actually have it! In other words, out of the 21 people in 1000 who would test positive, only 1 in 1000 would have the disease. Actually, this is not as bad as it initially looks: it's better, after all, to have a false positive than a false negative, as the false positive can be detected in further tests, whereas a false negative mandates no further tests. Also, the test does increase each of the 21 people's chances of having the disease from 1 in 1000 to 1 in 21.

At least, I'm 99.9999999999999% sure of that!

Okay, Back to Serious Stuff

I'm spending too much time with that juvenile LOLcats. Well, I'm middle-aged, what do you expect? I need laughs in my decrepitude - especially considering how decrepit my sense of humor is, anyway.

I'm still reading The Extended Phenotype. By my bones, I had another weird dream about it. (I've been having lots of weird dreams, such as Christopher Hitchens not liking my photos of the trip, me finding albatross eggs in my work desk, etc.) And it's becoming more and more difficult to summarize a text that is written with such economy and clarity - you may as well try to summarize poetry! ;-) I'm working on it, though.

Plus I've had an idea for a scholarly paper that I want to submit to one of the library journals. And you know what that means: I've put it off this long, but I cannot delay or deny it any longer. Yes. STATISTICS! What fun!

Actually, it is quite fun. That and calculus I am determined to conquer. (And The Extended Phenotype. And getting my fiction published.)

How are you spending your summer?

Gödel, Eschy, Blech

Sal Cordova inaction again.

Shimmies to Growth Rate nlgn.

UPDATED: MarkCC also takes down Cordova again at Good Math, Bad Math. Over and above the call of shimmies, Mark - shimmies and snake arms!

Computational Enzyme Kinetics and Proteomics (truncated)

Jonathan Vos Post, in response to my and others' requests, has kindly posted one of his papers at the Sal Cordova thread at Good Math, Bad Math.

(Assuming I can handle this level of discussion, I'll have to read it later this week because--ta da!--school starts for me again, and I've been busy reading one of my texts for the first night of classes. Everything you never wanted to know about indexing, abstracting, the construction of thesauri, etc. Doesn't that sound like fun? You know what's really scary is, it does to me. I'm having a ball reading this stuff... Maybe I could be a professional indexer. Augh! Nerd alert! Run!)

(And shimmy while you run! I'd like a little sugar for a change. *wink*)