Hello Cathy
It seems as though you may benefit through learning about ROC curves. There is a good tutorial provided on SPSS about their use and application.
Version 14.0 provides two very useful real-life examples by way of explaining how to geneate the relevant output and interpret it. You are also directed to the relevant datasets, hivassay.sav and bankloan.sav, which come with the package and which you should find by following the path: 'Program Files --> SPSS --> Tutorial --> sample_files' within the main directory of your computer.
To access the tutorial itself, go the menu 'Help' within SPSS and select 'Topics'. For your keyword search, type in 'ROC Curves' and scroll down the page to which you are directed until you see 'ROC Curves' under 'Related topics'. Click on this link and then on 'Show me' and allow the blocked content. The first example covers generation of a single ROC curve but the second example covers generation of multiple ROC curves on the one plot.
By the way, next time you write to the list, don't forget to include a subject header. This should increase your chance of receiving a reply.
Best wishes
Margaret
Cathy Fleming <
[hidden email]> wrote:
I am relatively inexperienced in SPSS and I have run across a problem.
The variable is the % of an organ the receives a particular radiation dose
and this variable is a significant predictor of side-effects in a Cox
regression. What I want to find is the cut-off point within this scale
variable that best divides patients into low and high-risk groups. I know
in theory that this could be done "by hand" but there are 275 entries, is
there any way of doing this automatically?
Thanks
Cathy
---------------------------------
The all-new Yahoo! Mail goes wherever you go - free your email address from your Internet provider.
Locked
