[R] Lack of independence in anova()

[Douglas Bates]

On 7/4/05, Douglas Bates <dmbates at gmail.com> wrote:
...
> The sums of squares for the different terms in the model are the sums
> of squares of elements of the effects vector as indexed by the assign
> vector.  The residual sum of squares is the sum of squares of the
> remaining elements of the assign vector. 

I meant to write "effects vector" not "assign vector" in that last sentence.