Close Home Forum Sign up / Log in

stats help again :-(

S

hi Sneaks
do you know how much you have helped me! thanks a lot. this means my study is under-powered :-(

I am not sure about another thing: using mean-centred scores in the regression; when we put into regress them, why do we use mean centred scores for predictors? And if we use mean-centred scores in the predictor, the dependent variable is not mean-centred. Why is that.

thanks very very very much
satchi

======= Date Modified 17 Mar 2010 12:30:15 =======
I never use mean-centred scores, I think its a form of transformation and would therefore only use them if my data is non-normal. I would assume if you do it to one variable then you should do it to them all (outcome AND predictor variables).


Anyone feel free to contradict me though!

I think they are often used to standardise the data.

S

hi sneaks
my friend just sent me a paper on mean-centering; would you like to see it; you'll probably understand it better than I do. PM me?
satchi

14258