Not an answer to your question, but what about Kendall's tau-a and tau-b,
Goodman and Kuskalls's gammbda and lambda .. . can't be bothered to get out
of my chair ...., and then there's the folk who use beta to mean
standardised regression slope. And if mu is the pop.mean, why not m for the
estimate?
----- Original Message -----
From: "Richard John Atkins" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Friday, October 03, 2003 9:37 AM
Subject: Spearman's ... rho?
> A trivial question but it's bugged me for a while. My understanding of
> terminology is that latin lower case letters tend to refer to statistics
> while greek lower case tends to refer to population parameters. While I
> appreciate statistical notation can be unconventional, I have often
> wondered why there are statistics such as Spearman's rho and Cohen's kappa
> which have greek letters.
>
> Did the latin letters run out?
> Is this a sign that Spearman and Cohen were free thinkers?
> Is there something systematic going on here?
> Is there a reason or a story behind this ... or do I simply need to get
out
> more?
>
> Richard Atkins
> Thames Valley University
|