[R-sig-Epi] Reference group in logistic regression analysis

[BXC (Bendix Carstensen)]

You presumably have your genotype variable coded as a factor; check by saying:

> str( gentp )

(or whatever your variable is called)
 
The default in glm is to use the first level as reference. But you can reorder the levels usuing the relevel() command, so you just do:

> glm( y ~ relevel(gentp,3) + ...., familily=binomial )

You can explore furtehr by trying:

> str( relevel(gento,3) )

In Epi there is slightly more versatile function called Relevel, which also allows

 > str( Relevel(gento,c(3,2)) )

which in some sence would be more natural.


Best regards,
Bendix
_________________________________________

Bendix Carstensen 
Senior Statistician
Steno Diabetes Center A/S
Niels Steensens Vej 2-4
DK-2820 Gentofte
Denmark
+45 44 43 87 38 (direct)
+45 30 75 87 38 (mobile)
bxc at steno.dk    www.biostat.ku.dk/~bxc
www.steno.dk


-----Original Message-----
From: r-sig-epi-bounces at stat.math.ethz.ch [mailto:r-sig-epi-bounces at stat.math.ethz.ch] On Behalf Of ????????? ??????????
Sent: 28. oktober 2010 09:57
To: r-sig-epi at stat.math.ethz.ch
Subject: [R-sig-Epi] Reference group in logistic regression analysis

To someone who could help me:

I am a PhD student, working in the Novosibirsk University, Russia. I am now using glm() function to do logistic regression analysis.
I cannot understand how the chosen reference group. For example, for the three genotypes of GG, GC, CC I want to be risk assessed in relation to the genotype CC, but the program selects the genotype GG. How do I change the reference group?


Thanks for your time.
Best wishes,
Ekaterina Kudryavtseva
PhD student,
Department of Molecular Biology
Novosibirsk University
Tel: +7-923-185-82-47

_______________________________________________
R-sig-Epi at stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-epi