Appologies, a minute ago I asked whether dichotomizing into quartiles is better for covariate variables than continuous variables:
another example is age: you cannot adjust a 10 year old as if they were 40 years old, therefore, it might be more useful to say: 0-10 (Y/N), 11-20 (Y/n) etc. Than it will to adjust for age as a continuous variable? Or will adjusting for age as a continuous variable have the same effect? . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
