Dear Colleagues,
How do we define a "potential effect modifier" in a model ?
How does one test for its presence in a model(e.g. logistic)? Should
one worry about its presence and if yes how can one control it ?
I am sure an effect modifier variable differs from a confounder
variable which is related to both the response(condition) and
exposure.
Could anyone please clarify?
Always thank you!
June E. Simakani
**********************************************************************
June Elijah Simakani, MSc, GradDip(Stats), DTE, C.Stat
Department of Statistics
University of Transkei
Private Bag X1, Unitra
Umtata 5117, South Africa
Tel.no: +27(47) 502 2529(work)
+27(47) 532 6380(res)
Fax.no: +27(47) 502 2525
**********************************************************************
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|