Hi
On Tue, 10 Feb 2004, sali wrote:
> I have 8 lines created from x,y scatters, I would like to
> test whether the slope is significantly different from line
> to line. How do I do this? (I use STATA, exel and sigma)
Do they come from 8 independent samples of x,y observations? If
so, are the 8 samples similar or different in various ways that
can be encoded by meaningful indicator variables for regression
purposes? If so, something akin to planned comparisons (i.e.,
the indicators) might be better than the 28 "pairwise"
comparisons of slopes that would be possible. For example,
Group Age Gender Ethnicity
1 -1 -1 -1
2 -1 -1 +1
3 -1 +1 -1
4 -1 +1 +1
5 +1 -1 -1
6 +1 -1 +1
7 +1 +1 -1
8 +1 +1 +1
would give you the "main effects" of age, gender, and ethnicity,
and their products with X (ideally centered by subtracting the
mean) would test differences of slopes by Age, Gender, and
Ethnicity. Higher order terms could also be computed,
corresponding to the 2- and 3-way interactions.
Other contrasts may be more meaningful given your design (e.g.,
linear, quadratic, and cubic predictors if Age had 4 levels).
If the 8 samples do not represent independent samples, then the
problem becomes even more challenging (I think).
Best wishes
Jim
============================================================================
James M. Clark (204) 786-9757
Department of Psychology (204) 774-4134 Fax
University of Winnipeg 4L05D
Winnipeg, Manitoba R3B 2E9 [EMAIL PROTECTED]
CANADA http://www.uwinnipeg.ca/~clark
============================================================================
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
. http://jse.stat.ncsu.edu/ .
=================================================================
