On Tue, 4 May 2004, Chen G wrote:
> I have 3 groups of usertypes marked "a", "b", and "c" that represent
> different psychological personalities that learn in the same class. In
> each group I have a different number of students ("a" - 13, "b" - 7,
> "c" - 7) and each have a grade.
> I would like to be able to tell whether there is a correlation (in
> the literal sense) between the different usertypes, and their grades.
> How do I do that statistically?
Probably the cleanest approach is to ask whether the mean grade differs
for the several groups. The standard analysis for questions of this
type is "analysis of variance". In particular, a one-way analysis of
variance.
(Using "correlation" in the usual statistical sense is possible, and
would be easy if you only had two groups; with three or more groups
that gets pretty messy, and I wouldn't recommend it for a novice.)
Consult any standard elementary statistics textbook. The calculations
are not difficult. There would however be some advantages to using a
statistical package for the analysis; MINITAB in particular has a nice
display of the several means, that makes it easy to see where the
differences are (if there are any).
------------------------------------------------------------
Donald F. Burrill [EMAIL PROTECTED]
56 Sebbins Pond Drive, Bedford, NH 03110 (603) 626-0816
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
. http://jse.stat.ncsu.edu/ .
=================================================================
