Hi
@JB1975,
re your earlier
Post #:74105258 (link here)
& your ongoing very concerned comment above " ...
And you can see that the p value under the two are different - you can check I didn't put any of those things - the similarities and the differences on the source documents they are still there.
From there you could ask simply why?Why are the dot patterns the same and the p values different?... "
& your further very gracious comment "
I am very well aware that that is way more forensic than almost anyone here will follow - but it is the data."
OK, you got me worried. Real worried. Maybe MSB's patents have a Problem (with a capital p-value).
Now, this appears to be the latest conspiracy theory to be announced by non-holders on HC:MSB. And they're sure coming thick & fast, along with the shorters. So, although I'm
not a physicist, or even a biostatistician, I'm going to have a crack at solving this one by doing nothing more than using a technique which I call 'looking at the pictures you posted'.
Ahem...OK, I just looked.
First - Spearman's rank correlation coefficient or rho (
r) has the little p-value calculated & displayed below it on the first diagram. It's written right there underneath it, viz:
Second - Coefficient of determination (
R2), which just for the heck of it I'm going to call 'R Squared', has the little p-value calculated & displayed on the second one. Also printed right alongside it, viz:
You're not going to believe this, but I have discovered that these things - Spearman's rank correlation coefficient & the Coefficient of Determination are actually two (2) different things.
Amazing. Huh. Apparently there is something to this statistics gig, after all. I'm a little curious - does anyone actually use statistics thingies over on your CYP stock?
Both these measures are effect size measures, apparently. I understand that Spearman's rho, for example, represents the degree of correlation of the data after data has been converted to ranks. It's said to capture the strength of relationship, you know, when one is assessing how well the relationship between two variables can be described using a monotonic function. Intuitively, the Spearman correlation between two variables will be
high - like 0.76 - when observations have a similar rank between the two variables, and low when observations have a dissimilar rank between the two variables. Whereas I heard some biostatisticians think the Coefficient of Determination is the proportion of the variation in the dependent variable that is predictable from the independent variable(s). Go figure.
Look, it's a simple mistake. Anyone could have made that mistake. Anyone.
Don't kick yourself.
@Pledge is here somewhere.
Cheers
GLTA(LT)H
(20min delay)