Dear All,
I am looking at a repeated-measures ANOVA which analysed the
orthogonal polynomial contrasts to assess the post-treatment growth
trend component of responses where data was collected one year
pre-treatment and after each growing season for years 1, 2, 3, 5,
and 10 post-treatment.
The researcher derived orthogonal polynomial coefficients (mean,
linear and quadratic) with the SAS/IML function ORPOL.
My questions are:
1) Is contr.poly(c(1, 2, 3, 5, 10), contrasts=T) the correct S-Plus
equivalent of the coefficients ORPOL would generate?
2) If so, how do I use/interpret the components:
> attributes(x)$coefs$alpha
[1] 4.20 7.16 4.25 2.99
> attributes(x)$coefs$norm2
[1] 1.0 5.0 50.8 300.5 627.6 305.4
I understand the linear, quadratic and higher order polynomials
to define trends with increasing number of inflection points
(i.e. linear: 0 (straight line), quadratic: 1 (U or inverted U),
cubic: 2, etc) but,
2) Does the mean polynomial coefficient compare the overall
mean across ALL years for each treatment? the intercept?
Many thanks in advance, DaveT.
**********************************************************
David J. Thompson
Silviculture Data Analyst
Ontario Forest Research Institute
Ontario Ministry of Natural Resources
1235 Queen Street East
Sault Ste. Marie, Ontario, P6A 2E5
(705) 946-7433
(705) 946-2030 Fax
david.thompson@mnr.gov.on.ca
**********************************************************
|