[R] Effective degrees of freedom

[Mario A. Pardo]

Greetings,

I am performing a simple Pearson's correlation test. Length of both 
vectors is 40, therefore the resulting df is 38. Nevertheless, a 
colleague is asking me for the "effective degrees of freedom". As far as 
I understand, those degrees of freedom have to be estimated for more 
complex regressions, but I was not able to find detailed information 
about it. Does any one of you know if I have to estimate those effective 
degrees of freedom for a simple Pearson correlation test?

Thank you very much in advance,

-- 
-mario