I'm holding a talk this afternoon about measuring heart pulse (that expansion of the artery you feel when you put your finger on your throat/wrist). I guess that any medical doctor will see that I know fuck all about arterial blood flow. Any electrical engineer will see that I know fuck all about electronics. And any physicist will see that my error bars are way too big.
I can see it, as I'm a physicist. You don't need to know anything about the subject to say that this graph is bullshit. Nature don't work like that.
If you're not a physicist, let me tell you a secret. If those error bars show one standard deviation, then if you fit a line through all the points, about one third of the points shouldn't touch the line with their error bars. A plausible plot looks something like this:
I didn't make the plots, and couldn't be bothered to sort out the error analysis. I'm just the gimp who does the talk. My worst case scenario is to tender my resignation on the spot (with 3 months' notice), go for a juggle in central park, and find a good strip club.
2 comments:
I wouldn't be sure that a physicist would notice... Most don't. That's what scares me.
Somehow I'm reminded of http://pages.cs.wisc.edu/~kovar/hall.html
Post a Comment