1 and 3 are the same model. Calling a variable a nuisance factor is just a description. The model is actually y = age*meditation*b1 + meditation*b2 + err = y_hat + err
doug
On 8/13/13 6:12 PM, ye tian wrote:
Dear Freesurfers,
I am studying cortical thickness as a function of age (continuous) and meditation practice (categorical, level1=Normal, level2=TM). I have three separate cases of qdec, regression1, regression2 and regression3.
Regression1: Discrete (Fixed Factor) = meditation; Continuous (Covariate) = age Regression2: Discrete (Fixed Factor) = meditation; Regression3: Discrete (Fixed Factor) = meditation; Nuisance Factor = age
For the question "Does the average thickness differ between Normal and TM?", Regression1 and Regression3 have identical result, which is different from Regression2, i.e., (1 = 3 !=2 ). May I infer from this result the following about qdec:
Regression1 and Regression3 both fit cortical thickness,y, to the model y = b0 + age*b1 + meditation*b2 + err = y_hat + err Regression2 fits: y = b0 + b2*meditation + err = y_hat + err
Some function of y_hat goes into the "average thickness", hence
the difference between 2 and the others.
However, what role does "Nuisance Factor" play in this case?
Thank you very much!
Sincerely, Ye
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer