good morning
i am running multinomial logistic regrssion and i have the folowing
question
i have a dependent variable with 4 level - the 4th being the reference
2 explanatory variables (coded 2,3,4(reference) and 0,1 (1 being the
reference) respectively
in the output for the 3rd level of the dependent (i could have used level
1 or level 2)
i get the following
var 1 coded 2 1.74
var 1 coded 3 (i have not given this figure so as not to confuse)
var 2 coded 0 1.04
when i calculate the effect of var1 and var 2 i get approx 1.81 (if
variable 2 is coded 0) and 1.67 (if variable 1 is coded 1)
if i stratify on variable 2 and repeat the initial analysis just including
variable 1 i get approx 1.64 and 1,90 repectively
can anyone explain why when i include the second variable in the equation
i get different answers as to when i stratify - i am particulary
interested in all of my results in relation to variable 2 coded 1 -
any comments
regards
ann
----------------------
ann e fitzmaurice
medical statistician
dugald baird centre
department of obs and gynae
university of aberdeen
[log in to unmask]
tel 01224 553876 (direct)
fax 01224 404925
|