complexitystudies » statistics

On the safety of flying

guenther — Sat, 21 Jul 2007 11:29:32 +0000

The post linked to below is actually about the safety of space shuttles, but I find another conclusion more interesting.

In the news you often hear that flying is safer than all other forms of travel, and this is true on a per kilometer basis. But not on a per hour basis (passenger hours vs. passenger kilometers) where air travel is indeed slightly more dangerous. (The discrepancy results from planes travelling more kilometers per hour than a car, of course.)

Want to get from point A to point B in one piece? Don’t take the shuttle.: Sciam Observations

So – getting from point A to B is safer by plane than by car – because the distance is fixed. But if you’re given the option between spending 8 hours in a plane or 8 hours in a car, the latter is safer.

Technorati Tags: statistics

Significance testing – mostly useless

guenther — Sun, 24 Jun 2007 20:56:50 +0000

One of my favourite pet topics is criticism of the current practice of statistical significance testing (which I found ridiculous from my first exposure to it). Here some reading on the feud:

Is significance testing all bad?

Technorati Tags: statistics

Error bars

guenther — Fri, 30 Mar 2007 11:34:14 +0000

To analyze data and also to critically review studies made by other people, one needs a good grasp of some basic concepts of statistics (and probability, but that is for another day). Here is a nice post about error bars: Cognitive Daily: I’ll bet you don’t understand error bars (updated with answers)

Technorati Tags: statistics

Statistical Significance

guenther — Tue, 20 Mar 2007 19:51:56 +0000

This is a nice thing to keep in mind (often forgotten):

Statistical significance means only that the pattern of distribution and relationship between variables which is found in the data from a sample can be confidently generalized to the larger populattion from which the sample was randomly drawn. By itself, it does not ensure that the relationship is theoretically or practically important or even very large.

From:

Chi Square Tutorial

Technorati Tags: statistics