s-news
[Top] [All Lists]

Re: factor levels

To: "natalie nicholls (IAH-C)" <natalie.nicholls@bbsrc.ac.uk>
Subject: Re: factor levels
From: Frank E Harrell Jr <f.harrell@vanderbilt.edu>
Date: Fri, 25 Feb 2005 10:22:35 -0500
Cc: s-news@lists.biostat.wustl.edu
In-reply-to: <8975119BCD0AC5419D61A9CF1A923E951BB682@iahce2knas1.iah.bbsrc.reserved>
References: <8975119BCD0AC5419D61A9CF1A923E951BB682@iahce2knas1.iah.bbsrc.reserved>
User-agent: Mozilla Thunderbird 0.9 (X11/20041105)
natalie nicholls (IAH-C) wrote:
Dear users,
In coxph - what does S use as the reference level in its output:
For example:

coef exp(coef) se(coef) z p V151 3.263 26.136 24.427 0.1336 0.89
V152  1.004     2.729    8.149  0.1232 0.90
V153  0.227     1.255    4.080  0.0556 0.96
V154  0.136     1.146    2.452  0.0556 0.96
V155 -0.772     0.462    3.989 -0.1935 0.85
V156  0.188     1.207    1.279  0.1471 0.88
V157  0.204     1.226    0.959  0.2127 0.83
V158  0.340     1.405    0.745  0.4568 0.65

where V151 etc, are different levels of a factor. I am looking at the effects 
of Year of Birth, the levels are:

"1991", "1992", "1993", "1994", "1995", "1996", "1997", "1998", "1999"

so is Splus comparing each year to 1991 (so V151 is 1992), or to 1999 (so V151 
is 1991), or just to each previous year in the list (so V151 is 1992, and is 
compared to 1991, with V152 being compared to V151 anad so on)?

Natalie

Please read the posting guide; show your code. It would also be good to use intuitive category names.

You might look at the Design library which has a wrapper function for coxph called cph. The output makes the coding apparent. cph uses the first level as the reference cell. For many things you don't need to know that (e.g., contrast, summary, plot, nomogram, latex functions in Design).

--
Frank E Harrell Jr   Professor and Chair           School of Medicine
                     Department of Biostatistics   Vanderbilt University

<Prev in Thread] Current Thread [Next in Thread>