Hi!
My question is on a test to compare CVs. The CVs are computed using the
same data but two different variance methods and I have to compare them.
Been told there is no real test and as of yet have not checked the Current
Index of Stat books but wondered if someone in the group has had this
problem. Someone suggested that take one of the Cvs and make it the
population CV and do 95% C.I. around that. Any suggestions?? Thanks.
-----Original Message-----
From: MJ Ray [mailto:[EMAIL PROTECTED]]
Sent: Sunday, October 29, 2000 7:46 AM
To: [EMAIL PROTECTED]
Subject: Re: FAVOUR - Marking Scripts Statistically!
[EMAIL PROTECTED] writes:
> I expected this to be a fairly common problem and hoped that a
> standard "formula" would already exist for it. As you quite correctly
> show, there are a huge number of possible scores given 10 grades (I
> thought of working this out but I think the factorials get too high for
> this time of night (01:50 GMT here in the UK).
This is, indeed, a fairly common problem. I think there are two
approaches you can take to it: the first is to decide that you as a
research student are too far away from the grade auditing system to
have a significant input *or even effect* on it, mark according to
whatever broken scheme you are given and hope "it all comes out in the
wash", that the combination of several monumental errors and
"adjustment" after auditing will cancel each other out and give
something approaching an accurate mark.
The second, which you are starting off on, is to decide that you
cannot mark assignments in such an unethical fashion. In this case,
you can either push for the scheme to be changed by whoever set it, or
unlaterally "redraft" the scheme and hope that the original author is
too overworked/uncaring to notice/care.
Either way, keep notes and remember your experiences when you come to
draw up your own mark schemes and systems. I think I favour setting
general principles and giving the marker significant discretion in
adjusting the scheme to make a fair and reasonable system. It's not
usually enough of a weighting in the degree to lose sleep over, unless
it happens for everything.
Finally, I feel obliged to say that these are my personal views and do
not necessarily reflect the opinion or policy of my University,
department or centre, as I can almost sense the knives being drawn as
I type.
MJR
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
