Albert le Curieux writes:
>I have to teach to med students a course on survival analysis with Excel
>only. Does somebody know some Excell macros I can use for testing and graphs.
The phrase "with Excel only" either implies that you are being forced to use Excel or
you are unable to use anything better. If it is the former case, protest vehemently
about this restriction. Medical students need Excel about as much as Accountants need
a stethescope.
If you do not have access to any software better than Excel, then I would suggest that
you ask every student to bring a pocket calculator. Anything will be fine as long as
it has a square root button (and you might not even need that). Show a very simple
example and ask the students to calculate a Kaplan-Meier curve by hand and free sketch
the curve on a sheet of paper. Get them to interpret values like the median survival
by drawing a horizontal line and the two and five year surivival probabilities by
drawing a vertical line.
If you were really, really ambitious, you could ask the students to compute some
standard errors as well. I'm not sure how messy the formulas are for the standard
error.
If you wanted to, you could use Excel as a fancy pocket calculator. I do that all the
time. But there is something to be said for getting closer to the data by using a
pocket calculator. Also, with a pocket calculator, you can have the class anywhere,
not just in a computer lab.
Then, rather than ask students to do a formal test like the log rank test or a Cox
regression model, give them some examples of how thes models are used in the published
literature. There are hundreds of good examples in journals with full text free on the
web.
Your typical medical student is unlikely to have to compute a Cox regression in
his/her future career, but will probably have to read a lot of journal articles that
talk about Cox regression.
For what it is worth, I have some pages that talk about the basics of survival
analysis. These pages use SPSS, but you might still find some of the material useful.
http://www.childrens-mercy.org/stats/model/survival.asp
http://www.childrens-mercy.org/stats/model/srv_data.asp
http://www.childrens-mercy.org/stats/model/kaplan.asp
There is some redundancy in the last page which I have not had time to fix yet.
Steve Simon, [EMAIL PROTECTED], Standard Disclaimer.
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
. http://jse.stat.ncsu.edu/ .
=================================================================
