Contrast or Dummy Variable Matrix
DESCRIPTION:
Returns a matrix giving contrasts or dummy variable coding.
USAGE: 
contr.sum(n, contrasts=T)
contr.poly(n, contrasts=T)
contr.helmert(n, contrasts=T)
contr.treatment(n, contrasts=T)
REQUIRED ARGUMENTS:
n:
a vector of levels for a category, or the number of levels. In the case of contr.poly, this can be the numeric values over which the contrasts should be orthogonal.
OPTIONAL ARGUMENTS:
contrasts:
should the contrasts be computed? If FALSE, dummy variables will be returned instead.
VALUE:
if contrasts is TRUE, a matrix with n rows and n-1 columns, and an n by n matrix if contrasts is FALSE. The interpretation of the matrix depends on the function:

contr.sum produces coefficients satisfying the constraint that their sum is zero (the traditional analysis of variance parametrization).

contr.poly creates orthogonal polynomials of degree 1, 2, etc., either on equally spaced points if n was a single number, or on the points specified by x. Columns are scaled to have norm 1.

For contr.helmert, contrast i is the difference between level i+1 and the average of levels 1:i.

In contr.treatment the contrasts are just levels 2 through n of the category.

DETAILS:
Note that contr.poly and contr.helmert produce orthogonal contrasts, contr.sum produces non-orthogonal contrasts, and contr.treatment produces a coding that is not technically a set of contrasts at all.

These functions are usually not called directly, but supplied as the coding for a category in a data frame.

To set the default contrast coding, change the contrasts option.

SEE ALSO:
contrasts , C , options .
EXAMPLES: 
# set contrasts to polynomial on x0 (a global dataset)
contrasts(my.vble) <-
        function(n,contrasts=F)contr.poly(x0,contrasts)


options(contrasts=c("contr.treatment", "contr.poly"))