Galen Wilkerson <[EMAIL PROTECTED]> wrote in
sci.stat.edu:
>Suppose you have two binary random variables, and you get some data
>from them:
>
>X Y
>----
>0 0
>1 0
>1 1
>1 1
>1 1
>1 1
>1 1
>0 0
>
>When you are calculating the correlation coefficient of these two, do
>you count the multiple occurances of the (1,1) and (0,0) pairs, or
>just use them once?
You count the multiple occurrences, just as you would with any other
data. Otherwise you are saying that every set has the exact same
correlation as this one:
0,0 0,1 1,0 1,1
>Please respond by email.
As Frank Slootweg <[EMAIL PROTECTED]> said in
comp.os.msdos.programmer on 27 Nov 1998:
"please email" requests are considered bad Netiquette (because
they imply that the poster is too lazy to check for possible
answers to his/her own questions, they waste the time of all
but one of the respondents, and they leave other users in the
dark). This makes that many/most people often do not respond
to "please email" questions, which obviously is the *poster's*
loss. Because I assume you mean well, and I am a nice guy :-),
I am posting this response anyway, and will e-mail you a copy.
Of course there are some valid reasons for asking for an
e-mailed *copy* of a (posted) response, like fast(er)
notification, (too) short expiry times for News articles,
etc.. If you have such a valid reason, then I suggest to just
say so in future, in order to prevent "lectures" in Netiquette
like this one.
More succinctly, quoting Herb Sutter in comp.lang.c++.moderated
on 7 Jun 1999:
if you don't have time for the group, why should the group
have time for you?
--
Stan Brown, Oak Road Systems, Cortland County, New York, USA
http://OakRoadSystems.com/
Support our troops: bring them home!
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
. http://jse.stat.ncsu.edu/ .
=================================================================
